WO2019142517A1 - Printing apparatus, and device, method, and program for detecting image defects - Google Patents

Printing apparatus, and device, method, and program for detecting image defects Download PDF

Info

Publication number
WO2019142517A1
WO2019142517A1 PCT/JP2018/044396 JP2018044396W WO2019142517A1 WO 2019142517 A1 WO2019142517 A1 WO 2019142517A1 JP 2018044396 W JP2018044396 W JP 2018044396W WO 2019142517 A1 WO2019142517 A1 WO 2019142517A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
defect
detection
defect candidate
unit
Prior art date
Application number
PCT/JP2018/044396
Other languages
French (fr)
Japanese (ja)
Inventor
正之 浮島
Original Assignee
富士フイルム株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 富士フイルム株式会社 filed Critical 富士フイルム株式会社
Publication of WO2019142517A1 publication Critical patent/WO2019142517A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis

Definitions

  • the present invention relates to a printing apparatus, an image defect detection apparatus, a method and a program, and more particularly to a technology for detecting an image defect from a printed matter.
  • Printed matter may cause image defects such as ink loss, ink dripping, scratches, streaks and the like during printing. For this reason, automatic inspection for detecting an image defect of a printed matter is performed.
  • a method of detecting an image defect there is known a method of imaging a printed matter with an imaging device to acquire an inspection image, and image processing and comparing a reference image acquired in advance and the inspection image. This image processing often uses less complex basic image processing. For example, after performing some basic preprocessing on the reference image and the inspection image, a differential image of the two images is obtained, and threshold processing is performed on the differential image to detect a signal exceeding the threshold as a defect.
  • preprocessing alignment of both images, spatial filter processing (blurring processing etc.) for reduction of noise components, or spatial filter processing (edge emphasis etc) for emphasizing a specific defect To go.
  • Patent Document 1 describes a technique for performing a detailed inspection of the entire surface of a printed matter at high speed.
  • a nozzle defect such as ejection bending or non-ejection occurs in a nozzle of an inkjet head
  • a stripe defect occurs in the printing position of the nozzle.
  • This streak defect is a very strong level because it is not printed at all at the printing position of the nozzle when the nozzle is not discharged.
  • the level is strong when the amount of curve is large, and the level is weak when the amount of curve is small.
  • parameter adjustment such as a threshold is often performed.
  • the parameter adjustment may be performed by the manufacturer or may be performed by the client via a UI (User Interface).
  • the level of image defects is ultimately determined by human visual judgment.
  • characteristics of the human visual system and the image pickup apparatus are different, it is not possible to sufficiently cope with basic image processing simply by parameter adjustment.
  • color tone characteristics are different between the human visual system and the imaging device, even if it is possible to appropriately set the presence or absence of a weak level and a strong level in a certain color tone, it is appropriate in another tone. It happens that it can not be set.
  • the background around the defect is recognized as a stronger level as it is closer to the flat network (the spatial frequency is lower), and as the background is more complex (the spatial frequency is higher) as a weaker level Do.
  • the present invention has been made in view of such circumstances, and it is an object of the present invention to provide a printing apparatus, an image defect detection apparatus, a method, and a program which are terminated within a target time while appropriately performing image defect detection processing. .
  • one aspect of the image defect detection apparatus includes an image acquisition unit for acquiring an inspection image obtained by imaging a printed matter and a reference image serving as a reference for detecting an image defect of the printed matter;
  • a first detection processing unit that detects a defect candidate position indicating a position of an image defect candidate from an inspection image, and a second detection process that compares the inspection image with a reference image to detect an image defect at the defect candidate position
  • the second detection processing unit is an image defect detection device that performs the second detection process only for the selected defect candidate position.
  • the inspection image obtained by imaging the printed matter and the reference image serving as the reference for detecting the image defect of the printed matter are acquired, and the defect detection positions indicating the positions of the image defect candidates from the inspected image are obtained by the first detection process.
  • the defect candidate positions to be subjected to the second detection process and having a number equal to or less than the upper limit are selected, and the inspection image and the reference image are selected only for the selected defect candidate positions. Since the image defect of the defect candidate position is detected by the second detection process of comparing the above, it is possible to finish within the target time while appropriately performing the image defect detection process.
  • the first detection processing unit detects the strength of the defect candidate position, the selection unit preferentially selects the defect candidate position where the detected strength is weak, and the defect candidate position not selected by the selection unit among the defect candidate positions It is preferable to include a determination unit that determines that an image defect exists. Thereby, the defect candidate position can be selected according to the detected intensity, and the defect candidate position not selected can be appropriately determined.
  • the first detection processing unit detects the strength of the defect candidate position, and the selection unit preferentially selects a defect candidate position having a high detected strength. Thereby, the defect candidate position can be selected according to the detected intensity.
  • the storage unit stores the defect candidate positions in the order of detection, and the selection unit selects the defect candidate positions in the normal order or the reverse order stored in the storage unit.
  • the selection unit selects the defect candidate positions in the normal order or the reverse order stored in the storage unit.
  • a setting unit configured to set presence / absence of notification by the notification unit is provided. This allows the user to know that the upper limit has been exceeded only when necessary.
  • the operation amount of the second detection process for the entire image is larger than the operation amount of the first detection process for the entire image. Further, it is preferable that the operation amount of the second detection process for a single area of the image is larger than the operation amount of the first detection process for a single area of the image.
  • the first detection process uses an inspection image and a reference image.
  • the defect candidate position can be appropriately detected.
  • the second detection process preferably uses an inspection partial image including a defect candidate position and a reference partial image corresponding to the inspection partial image. Thereby, the image defect of the defect candidate position can be appropriately detected.
  • the second detection process preferably includes a frequency analysis process. Thereby, detection corresponding to the human visual system can be performed.
  • the second detection process comprises a multiresolution analysis process. Thereby, detection corresponding to the human visual system can be performed.
  • the second detection process preferably includes determination by a machine learning model or a deep learning model. Thereby, detection corresponding to the human visual system can be performed.
  • the learning data for creating a machine learning model or a deep learning model preferably includes data obtained by processing sensory values or sensory values obtained by human sensory evaluation experiments.
  • the machine learning model or the deep learning model can express human visual characteristics, and detection corresponding to the human visual system can be performed.
  • the second detection processing unit preferably includes a parallel computing unit that processes a plurality of operations in parallel. Thereby, the second detection process can be performed at high speed, and the upper limit value can be increased.
  • the parallel computing unit comprises a plurality of CPUs (Central Processing Units).
  • the parallel computing unit comprises a plurality of substrates.
  • the parallel computing unit may include a plurality of GPUs (Graphics Processing Units). By configuring in this manner, the parallel computing unit can be appropriately configured.
  • one aspect of the printing apparatus is a printing unit for printing a printed matter, an imaging unit for imaging a printed matter to acquire an inspection image, and a reference image serving as a reference for detecting an image defect of the printed matter
  • a reference image acquisition unit to acquire a first detection processing unit that detects a defect candidate position indicating a position of an image defect candidate from an inspection image by a first detection process, and a second detection that compares the inspection image and the reference image
  • a second detection processing unit that detects an image defect at a defect candidate position by processing, a selection unit that selects a defect candidate position at which the second detection processing is performed among the detected defect candidate positions, and a detection result of the second detection processing
  • the selection unit selects defect candidate positions having a number equal to or less than the upper limit value, and the second detection processing unit determines the selected defect candidate positions. Only the second detection processing mark It is a device.
  • the printed matter is printed, the printed matter is imaged, the inspection image is acquired, the reference image serving as the reference for detecting the image defect of the printed matter is acquired, and the candidate for the image defect from the inspection image is obtained by the first detection process.
  • Detects the defect candidate position indicating the position of the defect candidate position selects the number of defect candidate positions that are the defect candidate positions to be subjected to the second detection processing among the detected defect candidate positions and not more than the upper limit, and selects the selected defect candidate positions Since the image defect of the defect candidate position is detected by the second detection process of comparing the inspection image and the reference image only for the above, the quality of the inspection image is judged based on the detection result of the second detection process.
  • the quality determination of the inspection image can be completed within the target time while appropriately performing the defect detection processing.
  • one aspect of the image defect detection method includes an image acquisition step of acquiring an inspection image obtained by imaging a printed matter and a reference image serving as a reference for detecting an image defect of the printed matter;
  • the second detection process is an image defect detection method in which the second detection process is performed only for the selected defect candidate position.
  • the inspection image obtained by imaging the printed matter and the reference image serving as the reference for detecting the image defect of the printed matter are acquired, and the defect detection positions indicating the positions of the image defect candidates from the inspected image are obtained by the first detection process.
  • the defect candidate positions to be subjected to the second detection process and having a number equal to or less than the upper limit are selected, and the inspection image and the reference image are selected only for the selected defect candidate positions. Since the image defect of the defect candidate position is detected by the second detection process of comparing the above, it is possible to finish within the target time while appropriately performing the image defect detection process.
  • One aspect of a program for causing a computer to execute the image defect detection method to achieve the above object acquires an inspection image obtained by imaging a printed matter and a reference image serving as a reference for detecting an image defect of the printed matter
  • a defect detection candidate detecting a defect candidate position indicating a position of an image defect candidate from the inspection image by the first detection processing, and a second detection processing comparing the inspection image with the reference image
  • a second detection process step of detecting an image defect at a position and a selection step of selecting a defect candidate position at which the second detection process is to be performed among the detected defect candidate positions
  • the inspection image obtained by imaging the printed matter and the reference image serving as the reference for detecting the image defect of the printed matter are acquired, and the defect detection positions indicating the positions of the image defect candidates from the inspected image are obtained by the first detection process.
  • the defect candidate positions to be subjected to the second detection process and having a number equal to or less than the upper limit are selected, and the inspection image and the reference image are selected only for the selected defect candidate positions. Since the image defect of the defect candidate position is detected by the second detection process of comparing the above, it is possible to finish within the target time while appropriately performing the image defect detection process.
  • one aspect of the image defect detection apparatus is an image defect detection apparatus having a processor, wherein the processor is a reference image that serves as a reference for detecting an inspection image obtained by imaging a printed material and an image defect of the printed material , And detects a defect candidate position indicating the position of an image defect candidate from the inspection image by the first detection process, and detects the number of defects equal to or less than the upper limit value for performing the second detection process among the detected defect candidate positions. It is an image defect detection device which detects an image defect of a defect candidate position only for the selected defect candidate position by a second detection process of selecting a candidate position and comparing an inspection image and a reference image.
  • Block diagram showing the internal configuration of the image defect detection apparatus Flow chart showing processing of image defect detection method Block diagram showing the internal configuration of the image defect detection apparatus Flow chart showing processing of image defect detection method Flow chart showing processing of image defect detection method Block diagram showing the internal configuration of the image defect detection apparatus Block diagram showing the internal configuration of the image defect detection apparatus Flow chart showing processing of image defect detection method Overall configuration of inkjet printing device Block diagram showing the internal configuration of the ink jet printing apparatus Flow chart showing processing of printing method and data used for each processing
  • FIG. 1 is a block diagram showing an internal configuration of the image defect detection apparatus 10 according to the first embodiment.
  • the image defect detection device 10 is a device that detects an image defect of a printed matter.
  • the image defect detection apparatus 10 includes an image acquisition unit 20, a first detection processing unit 22, a selection unit 24, and a second detection processing unit 26.
  • the image acquisition unit 20 acquires an inspection image obtained by imaging a printed matter for detecting an image defect. Further, the image acquisition unit 20 acquires a reference image which is a reference for detecting an image defect of a printed matter.
  • the image acquisition unit 20 may acquire an inspection image and a reference image from an imaging device (not shown) provided in the device, or may acquire it from an input interface (not shown).
  • the first detection processing unit 22 detects a defect candidate position indicating a position of an image defect candidate from the inspection image by the first detection process. Details of the first detection process will be described later.
  • the selection unit 24 selects, from among the defect candidate positions detected by the first detection process, a defect candidate position to be subjected to a second detection process described later.
  • the selection unit 24 selects the number of defect candidate positions in the range of the predetermined upper limit value or less.
  • the second detection processing unit 26 detects an image defect of the defect candidate position by a second detection process of comparing the inspection image and the reference image. Details of the second detection process will be described later.
  • FIG. 2 is a flowchart showing processing of the image defect detection method and data used for each processing.
  • step S1 the image acquiring unit 20 captures a reference image D1 serving as a reference for detecting an image defect and a printed matter for detecting an image defect.
  • An inspection image D2 is acquired.
  • the reference image D1 in the present embodiment is image data obtained by imaging the non-defective printed material printed based on the same input image data (print source data) as the printed material for which an image defect is detected, by the imaging device.
  • the image acquisition unit 20 reads out and acquires a reference image D1 stored in advance in a memory (not shown).
  • the image acquisition unit 20 may acquire the reference image D1 from an input interface (not shown). Further, the image acquisition unit 20 acquires the inspection image D2 from an input interface (not shown).
  • the reference image D1 is not limited to the image captured by the imaging device.
  • input image data of the printing apparatus when printing a printed matter may be used as a reference image.
  • input image data subjected to resolution conversion processing, color conversion processing, gradation conversion processing, screening processing or the like may be used as the reference image D1.
  • the acquired reference image D1 and inspection image D2 are input to the first detection processing unit 22 and the second detection processing unit 26.
  • step S2 an example of a first detection process
  • the first detection processing unit 22 performs the first detection process to detect the defect candidate position D3 of the inspection image D2.
  • the detected defect candidate position D3 is input to the selection unit 24.
  • step S3 an example of a selection step
  • the selection unit 24 selects the number of selected defect candidate positions D5 equal to or less than the upper limit value D4 from among the plurality of defect candidate positions D3.
  • the upper limit value D4 is a value predetermined in accordance with a target time for detecting an image defect.
  • the upper limit value D4 is stored in advance in a memory (not shown).
  • the selected selected defect candidate position D5 is input to the second detection processing unit 26.
  • step S4 an example of a second detection processing step
  • the second detection processing unit 26 performs a second detection processing based on the reference image D1 and the inspection image D2, and applies to each selected defect candidate position D5. Image defect D6 is detected.
  • the first detection process uses the reference image D1 and the inspection image D2.
  • the first detection processing unit 22 performs resolution matching processing of the reference image D1 and the inspection image D2, gradation matching processing, brightness matching processing, color matching processing, alignment processing, noise reduction processing as pre-processing of the first detection processing.
  • spatial filtering spatial filtering for enhancing image defects may be performed.
  • the first detection process is, for example, a process of generating a difference image between the reference image D1 and the inspection image D2 and setting a position where the absolute value of each pixel of the difference image exceeds a certain threshold as a defect candidate position.
  • the position where the pixel value of each pixel exceeds a threshold is the defect candidate position It may be a process to In this case, a defect candidate position brighter than the reference image D1 can be detected.
  • the pixel value of each pixel exceeds a certain threshold to be a defect candidate It may be processing to be a position. In this case, a defect candidate position darker than the reference image D1 can be detected.
  • the first detection process may be a process performed for each area. For example, first, the reference image D1 and the inspection image D2 are divided into grids, respectively. Next, a difference grid image is generated for each grid. Further, statistical values (average value, median value, etc.) of the absolute values of this difference grid image are calculated. Finally, this statistic is compared with a threshold. When the threshold value is exceeded, the center position of the grid corresponding to the difference grid image is set as the defect candidate position. Note that the division into grids may be performed with overlapping allowed.
  • the first detection process may be the following process.
  • the reference image D1 and the inspection image D2 are divided into grids, respectively.
  • a reference image profile is generated in which the reference image D1 is integrated for each grid in accordance with the direction of the streak defect.
  • an inspection image profile is generated in which the inspection image D2 is integrated for each grid in accordance with the direction of the streak defect.
  • a difference profile between the reference image profile and the inspection image profile is generated.
  • the intensity value of the difference profile exceeds a certain threshold
  • the lattice corresponding to the difference profile is set as a defect candidate position. Note that the division into lattices may be performed with overlapping allowed.
  • a plurality of these processes may be performed as the first detection process.
  • the first detection process is not limited to the above process, as long as the defect candidate position can be detected from the inspection image.
  • the second detection process performs an image defect close to human recognition.
  • the second detection process is, for example, the following process. First, for each of the selected defect candidate positions D5, a reference partial image and an inspection partial image which are partial images centering on the selected defect candidate positions D5 of the reference image D1 and the inspection image D2 are acquired.
  • the reference partial image and the inspection partial image are not limited to those centered on the selected defect candidate position D5, and may include the selected defect candidate position D5.
  • a difference partial image is generated from the reference partial image and the inspection partial image.
  • the feature amount of the area is calculated using at least a part of the reference partial image, the inspection partial image, and the difference partial image.
  • Whether the defect candidate at the selected defect candidate position D5 is the image defect D6 by inputting this feature amount into a machine learning model (multiple regression analysis, partial least squares method, neural network, support vector machine, ridge regression, etc.) Determine (classification) whether or not non-defective.
  • alignment processing of the reference image D1 and the inspection image D2 may be performed.
  • the second detection process may be a process of inputting the reference partial image, the inspection partial image, and the difference partial image into a machine learning model generated by deep learning, and detecting the image defect D6 by the machine learning model.
  • a base image is generated by performing a Fourier transform process, a frequency analysis process using wavelet transform, or a multiple resolution analysis process on the reference partial image, the inspection partial image, and the difference partial image.
  • the image defect D6 may be detected from the feature amount obtained from the generated base image.
  • a base image is generated by performing Fourier transform processing, frequency analysis processing using wavelet transform or the like, or multiple resolution analysis processing on the reference partial image, inspection partial image, and difference partial image.
  • the process may be a process of inputting the generated base image or a feature value obtained from the base image into a machine learning model or a deep learning model, and detecting the image defect D6 by the machine learning model or the deep learning model.
  • the image may be regressed to an amount matched to human visual characteristics, and the image defect D6 may be determined when the threshold value is exceeded for the regression result.
  • the model can well express human visual characteristics. For this reason, as learning data for creating a model, it is desirable to use sensory values obtained by human sensory evaluation experiments or data obtained by processing the sensory values.
  • the second detection process is not limited to the above process, as long as the defect candidate at the defect candidate position can be classified as to whether or not it is the image defect D6.
  • the second detection processing unit 26 may include a parallel computing unit (not shown) that processes a plurality of operations in parallel. By using the parallel computing unit, it is possible to execute the second detection process in parallel to speed up the process. As a result, the upper limit value D4 can be made a large value.
  • the parallel computing unit is configured to include, for example, a plurality of CPUs (Central Processing Units) mounted on a single substrate.
  • the parallel computing unit may be configured to include a plurality of substrates on which one or more CPUs are mounted.
  • CPUs Central Processing Units
  • the parallel computing unit may be configured to include a plurality of substrates on which one or more CPUs are mounted.
  • the parallel computing unit may be configured to include a plurality of GPUs (Graphics Processing Units) instead of or together with the CPU.
  • the first detection processing unit 22 and the second detection processing unit 26 may share the same parallel computing unit.
  • the second detection process performs a relatively complicated inspection, so the amount of calculation per unit inspection of the second detection process is This is larger than the amount of calculation per unit inspection of the 1 detection process. That is, the operation amount of the second detection process for the entire image is larger than the operation amount of the first detection process for the entire image. This means that if the second detection process is performed assuming that all pixel positions of the image are defect candidate positions, the amount of calculation is large compared to the first detection process.
  • the operation amount of the second detection processing for a single area of the image is larger than the operation amount of the first detection processing for the same single area.
  • a parallel computing unit is used for the second detection processing unit 26, it is preferable to compare the entire image.
  • the amount of operation per unit inspection of the second detection process is relatively large. For this reason, when the second detection process is performed on the entire image, the amount of calculation becomes enormous, and the calculation time of the image defect detection process does not fall within the target time.
  • the second detection processing unit 26 performs the second detection process only on defect candidate positions having a number equal to or less than a predetermined upper limit value D4 among the defect candidate positions detected by the first detection process. Thereby, the image defect detection process can be completed within the target time.
  • the upper limit value D4 used in the selection unit 24 may be the largest integer among the integers satisfying Expression 1.
  • the time T 1 is a time determined by the processing capability of the first detection processing section 22.
  • the time T 2 is a time determined by the processing capability of the second detection processing section 26.
  • the time T 1 and time T 2 are, may vary by the size and complexity of the image defect detection target image. In the present embodiment, the time T 1 is a value determined without depending on the image.
  • the upper limit value of the number of defect candidate positions on which the second detection process is performed, the target time of the detection time Since it can be set according to the processing capacity and the processing capacity of the second detection processing unit 26, it is possible to guarantee the implementation of the highly accurate second detection processing within the processing time.
  • the second detection process since the second detection process can be designed with priority given to performance over operation time, it is possible to make the second detection process complicated and have high performance.
  • FIG. 3 is a block diagram showing an internal configuration of the image defect detection device 12 according to the second embodiment.
  • symbol is attached
  • the image defect detection device 12 is a device that detects an image defect of a printed matter. As shown in FIG. 3, the image defect detection device 12 differs from the image defect detection device 10 in that the image defect detection device 12 includes a determination unit 28 and a quality determination unit 30.
  • the determination unit 28 determines a defect candidate position not selected by the selection unit 24 among the defect candidate positions detected by the first detection process.
  • the quality determination unit 30 determines the quality of the printed matter based on the detection result of the second detection processing unit 26 and the determination result of the determination unit 28.
  • FIG. 4 is a flowchart showing processing of the image defect detection method and data used for each processing.
  • symbol is attached
  • step S1 the reference image D1 and the inspection image D2 are acquired by the image acquisition unit 20.
  • step S2 the first detection processing unit 22 performs a first detection process to detect a defect candidate position D3 of the inspection image D2.
  • the first detection processing unit 22 detects the intensity value of the defect candidate position D3.
  • the intensity value of the defect candidate position D3 is an index indicating the visibility of the image defect candidate.
  • the intensity value of the defect candidate position D3 is relatively large as the intensity is high, that is, as the image defect candidate is easy to be recognized, and as the intensity is low, ie as it is difficult to be visually recognized.
  • step S11 the selection unit 24 selects the number of selected defect candidate positions D5 equal to or less than the upper limit value D4 from among the plurality of defect candidate positions D3.
  • the number of defect candidate positions D3 is 20 and the upper limit value D4 is 15.
  • the selection unit 24 preferentially selects the defect candidate position D3 whose intensity is weak. That is, the plurality of defect candidate positions D3 are sorted in the order of weak intensity, and the 15 first to fifteenth positions are selected as the selected defect candidate position D5 in ascending order of intensity.
  • the selected selected defect candidate position D5 is input to the second detection processing unit 26.
  • the selection unit 24 sets a defect candidate position other than the selected defect candidate position D5 among the plurality of defect candidate positions D3 as a non-selected defect candidate position D11. That is, five non-selected defect candidate positions D11 from the sixteenth to the twentieth places in the order of weak intensity.
  • the non-selected defect candidate position D11 is input to the determination unit 28.
  • the determination unit 28 determines that an image defect exists at the input non-selected defect candidate position D11.
  • the determination result D12 that the image defect D6 exists at five non-selected defect candidate positions D11 is output.
  • step S4 the second detection processing unit 26 performs a second detection process of detecting the image defect D6 with respect to 15 selected defect candidate positions D5.
  • step S13 the good or defective judgment unit 30 judges the quality of the printed matter based on the image defect D6 by the judgment result D12 of the judgment unit 28 and the image defect D6 which is the detection result of the second detection processing unit 26.
  • the pass / fail judgment result D13 is output.
  • the pass / fail judgment is performed with ten pass / fail judgment thresholds of the number of image defects D6. In this case, if the sum of the number of image defects D6 according to the determination result D12 of the determination unit 28 and the image defect D6 which is the detection result of the second detection processing unit 26 is less than 10, non-defective printed matter (accepted printed matter) If it is above, it will be judged as inferior goods printed matter (rejected printed matter).
  • the second detection process is performed by giving priority to defect candidate positions having weak intensities, and it is determined that an image defect exists at a defect candidate position where the second detection process is not performed. By doing so, it becomes easy to make subtle judgments that make it correspond to the human visual system as highly accurate defect detection. Therefore, it is possible to complete the process within the target time while properly performing the image defect detection process.
  • step S11 As a result of selecting the selected defect candidate position D5 in step S11, if the number of non-selected defect candidate positions D11 exceeds the pass / fail judgment threshold, it is immediately judged as a defective printed matter and the second detection process is omitted. Good.
  • FIG. 5 is a flowchart showing the process of the image defect detection method according to the third embodiment in the image defect detection apparatus 12 and the data used for each process.
  • the parts common to the flowchart shown in FIG. 4 are denoted by the same reference numerals, and the detailed description thereof is omitted.
  • step S1 the reference image D1 and the inspection image D2 are acquired by the image acquisition unit 20. Further, in step S2, the defect candidate position D3 of the inspection image D2 and its intensity (intensity value) are detected.
  • step S21 the selection unit 24 selects the number of selected defect candidate positions D5 equal to or less than the upper limit value D4 from among the plurality of defect candidate positions D3.
  • the number of defect candidate positions D3 is 20 and the upper limit D4 is 15.
  • the selection unit selects fifteen selected defect candidate positions D5.
  • the selection unit 24 preferentially selects the defect candidate position D3 having a high strength. That is, the plurality of defect candidate positions D3 are sorted in descending order of strength, and the 15 first to fifteenth positions are selected as the selected defect candidate positions D5 in descending order of strength.
  • the selected selected defect candidate position D5 is input to the second detection processing unit 26.
  • the selection unit 24 sets a defect candidate position other than the selected defect candidate position D5 among the plurality of defect candidate positions D3 as a non-selected defect candidate position D11. That is, five non-selected defect candidate positions D11 are the sixteenth to twentieth places in descending order of strength.
  • the non-selected defect candidate position D11 is input to the determination unit 28.
  • the determination unit 28 determines that there is no image defect at the input non-selected defect candidate position D11.
  • a determination result D12 indicating that no image defect exists at five non-selected defect candidate positions D11 is output.
  • step S4 the second detection processing unit 26 detects the image defect D6 at 15 selected defect candidate positions D5.
  • step S12 the good or bad judgment unit 30 judges the quality of the printed matter based on the image defect D6 which is the detection result of the second detection processing unit 26, and outputs the good / bad judgment result D13.
  • the second detection process is performed with priority given to the defect candidate position having high intensity, and the image defect does not exist at the defect candidate position where the second detection process is not performed.
  • the determination it is possible to complete the process within the target time while appropriately performing the image defect detection process.
  • it is effective in the case where it is determined as a defective product printed matter if an image defect exists even in one place.
  • the image defect detection device 12 may not include the determination unit 28.
  • the second detection process may be omitted by immediately judging as a non-defective printed matter.
  • FIG. 6 is a block diagram showing an internal configuration of the image defect detection device 14 according to the fourth embodiment.
  • symbol is attached
  • the selection unit 24 of the image defect detection device 14 includes a storage unit 32.
  • the storage unit 32 stores the defect candidate position detected by the first detection processing unit 22 at the end of the queue. Further, the selection unit 24 selects a defect candidate position from the head of the queue. That is, the selection unit 24 functions as a so-called queue that selects defect candidate positions in the order of being stored in the storage unit 32.
  • the selection unit 24 selects the defect candidate position D3 as the selected defect candidate position D5 in the order stored in the storage unit 32.
  • the remaining non-selected defect candidate positions D3 are set as non-selected defect candidate positions D11.
  • the selection unit 24 preferentially selects the defect candidate position D3 whose intensity is weak. Further, in the third embodiment, the selection unit 24 selects the defect candidate position D3 with high strength in priority.
  • the upper limit value D4 can be increased without performing prioritization.
  • the selection unit 24 may select the defect candidate position from the end of the queue. That is, the selection unit 24 functions as a so-called stack that selects defect candidate positions in the reverse order stored in the storage unit 32. Even with this configuration, since the prioritization is not performed, the upper limit value D4 can be increased.
  • FIG. 7 is a block diagram showing an internal configuration of the image defect detection apparatus 16 according to the fifth embodiment.
  • the same reference numerals as in the block diagram shown in FIG. 3 denote the same parts in FIG.
  • the image defect detection device 16 includes a notification unit 34 and a setting unit 36.
  • the notification unit 34 notifies the user of a warning or an error when the number of defect candidate positions exceeds the upper limit value.
  • the notification unit 34 may display an indication that the upper limit value is exceeded on the display unit (not shown), may issue a message from a speaker (not shown), or may turn on a light (not shown).
  • the setting unit 36 sets the presence or absence of notification by the notification unit 34.
  • the user can input the presence or absence of notification using an input interface (not shown).
  • the input information is set in the setting unit 36.
  • the notification unit 34 notifies the user when the notification is set in the setting unit 36 and the number of defect candidate positions exceeds the upper limit value.
  • FIG. 8 is a flowchart showing processing of the image defect detection method and data used for each processing.
  • the parts common to the flowchart shown in FIG. 4 are denoted by the same reference numerals, and the detailed description thereof is omitted.
  • step S31 the user inputs the presence or absence of notification by the notification unit 34 using an input interface (not shown).
  • the setting unit 36 sets the presence or absence of notification by the notification unit 34 based on the input information.
  • step S1 the reference image D1 and the inspection image D2 are acquired by the image acquisition unit 20.
  • step S2 the first detection processing unit 22 performs the first detection process to detect the defect candidate position D3 of the inspection image D2 and the intensity thereof.
  • the defect candidate position D3 is input to the notification unit 34.
  • the notification unit 34 acquires an upper limit value D4 stored in a memory (not shown). Furthermore, the notification unit 34 compares the upper limit value D4 with the number of defect candidate positions D3 input from the first detection processing unit 22, and determines whether the number of defect candidate positions D3 exceeds the upper limit value D4. judge.
  • step S33 If the number of defect candidate positions D3 exceeds the upper limit value D4, the process proceeds to step S33. On the other hand, when the number of defect candidate positions D3 does not exceed the upper limit value D4, the process proceeds to step S11 without performing the notification.
  • step S33 the notification unit 34 determines whether the setting unit 36 has set notification. If the notification is set, the process proceeds to step S34. Moreover, when it sets to no notification, it transfers to step S11, without performing notification.
  • step S34 the notification unit 34 notifies the user of a warning or an error that the number of defect candidate positions D3 has exceeded the upper limit value D4. Then, it transfers to step S11.
  • step S11 the selection unit 24 selects the number of selected defect candidate positions D5 equal to or less than the upper limit value D4 from among the plurality of defect candidate positions D3.
  • the subsequent processing is the same as that of the second embodiment.
  • the user can know that by notifying the user.
  • the image defect detection devices 10, 12, 14 and 16 can be applied to a printing device.
  • an inkjet printing apparatus to which the image defect detection device 12 is applied will be described.
  • FIG. 9 is an overall configuration diagram showing an overall schematic configuration of the inkjet printing apparatus 100. As shown in FIG. As shown in FIG. 9, the inkjet printing apparatus 100 ejects four color inks of cyan (C), magenta (M), yellow (Y), and black (K) onto a sheet of paper P as a printing medium. Printing machine that prints color images.
  • C cyan
  • M magenta
  • Y yellow
  • K black
  • a general-purpose printing sheet is not a so-called inkjet-only sheet but a sheet mainly composed of cellulose such as a coated sheet used for general offset printing.
  • an aqueous ink is used as the ink.
  • the aqueous ink refers to an ink in which a coloring material such as a dye or a pigment is dissolved or dispersed in water and a solvent soluble in water.
  • the inkjet printing apparatus 100 includes a conveyance unit 110, a printing unit 120, an imaging unit 130, a drying unit 140, a sorting unit 150, a paper discharge unit 160, and the like.
  • the transport unit 110 transports the sheet P fed from a paper feed unit (not shown) in the transport direction (Y direction).
  • the conveyance unit 110 includes an upstream pulley 112, a downstream pulley 114, and a conveyance belt 116.
  • the upstream pulley 112 has a rotating shaft (not shown) extending in the horizontal direction, and the rotating shaft is rotatably supported.
  • the downstream pulley 114 has a rotating shaft (not shown) parallel to the rotating shaft of the upstream pulley 112, and the rotating shaft is rotatably supported.
  • the conveying belt 116 is an endless belt made of stainless steel.
  • the transport belt 116 is stretched over the upstream pulley 112 and the downstream pulley 114.
  • the conveying belt 116 made of stainless steel By using the conveying belt 116 made of stainless steel, the flatness of the sheet P can be maintained well.
  • the downstream pulley 114 has a motor (not shown) as drive means. When the motor is driven, the downstream pulley 114 rotates counterclockwise in FIG. The upstream pulley 112 rotates counterclockwise in FIG. 9 following the rotation of the downstream pulley 114. The rotation of the upstream pulley 112 and the downstream pulley 114 causes the transport belt 116 to travel along the traveling path between the upstream pulley 112 and the downstream pulley 114.
  • the sheet P supplied from a sheet feeding unit (not shown) is placed on the conveyance surface of the conveyance belt 116.
  • the conveyance unit 110 conveys the sheet P placed on the conveyance belt 116 along a conveyance path from the upstream pulley 112 to the downstream pulley 114 and delivers the sheet P to the paper discharge unit 160.
  • the sheet P is conveyed with the printing surface held horizontally at a position facing the printing unit 120, the imaging unit 130, the drying unit 140, and the sorting unit 150 in the transport path.
  • a plurality of suction holes are provided in the conveyance belt 116, and the suction holes of the conveyance belt 116 are suctioned by a pump (not shown), whereby the sheet P placed on the conveyance surface of the conveyance belt 116 is adsorbed and held on the conveyance surface.
  • the printing unit 120 forms (prints) an image on the sheet P.
  • the printing unit 120 includes inkjet heads 122C, 122M, 122Y, and 122K.
  • the ink jet head 122C ejects ink droplets of cyan (C) by the ink jet method.
  • the inkjet heads 122M, 122Y, and 122K eject ink droplets of magenta (M), yellow (Y), and black (K) by an inkjet method, respectively.
  • the inkjet heads 122 C, 122 M, 122 Y and 122 K are disposed at regular intervals along the transport path of the sheet P by the transport belt 116.
  • the inkjet heads 122C, 122M, 122Y, and 122K are respectively line heads, and are formed to have a length corresponding to the maximum sheet width.
  • the inkjet heads 122C, 122M, 122Y and 122K are arranged such that the nozzle surfaces (surfaces on which the nozzles are arranged) face the transport belt 116.
  • the inkjet heads 122C, 122M, 122Y, and 122K form an image on the printing surface of the sheet P by discharging ink droplets from the nozzles formed on the nozzle surface toward the sheet P conveyed by the conveyance belt 116. .
  • the imaging unit 130 acquires an image of the printing surface of the sheet P.
  • the imaging unit 130 is disposed downstream of the printing unit 120 with respect to the conveyance direction of the sheet P.
  • the imaging unit 130 includes a scanner 132.
  • the scanner 132 is a device that optically reads an image formed on the sheet P using the inkjet heads 122C, 122M, 122Y, and 122K, and generates image data indicating the read image.
  • the scanner 132 includes an imaging device that captures an image printed on the sheet P and converts the image into an electrical signal.
  • a color charge coupled device (CCD) linear image sensor can be used as an imaging device.
  • a color CMOS (complementary metal oxide semiconductor) linear image sensor may be used instead of the color CCD linear image sensor.
  • the scanner 132 may include, in addition to an imaging device, an illumination optical system that illuminates an object to be read, and a signal processing circuit that processes a signal obtained from the imaging device to generate digital image data.
  • the drying unit 140 dries the ink of the paper P.
  • the drying unit 140 is disposed downstream of the imaging unit 130 in the transport direction of the sheet P.
  • the drying unit 140 includes a heater 142.
  • the heater 142 for example, at least one of a halogen heater and an infrared heater is used.
  • the heater 142 heats the printing surface of the sheet P to dry the ink of the sheet P.
  • the drying unit 140 may include blowing means such as a fan or a blower.
  • the sorting unit 150 sorts the printed matter according to the quality determination regarding the sheet P conveyed by the conveyance belt 116.
  • the sorting unit 150 is disposed downstream of the drying unit 140 in the transport direction of the sheet P.
  • the sorting unit 150 includes a stamper 152.
  • the stamper 152 performs a stamping process to attach the ink to the leading edge of the sheet P determined to be a defective printed matter according to the quality determination regarding the sheet P transported by the transport belt 116.
  • the paper discharge unit 160 collects the sheet P (printed matter) on which the image is formed and dried.
  • the paper discharge unit 160 is disposed downstream of the sorting unit 150 in the conveyance direction of the sheet P and at the end point of the conveyance path of the conveyance unit 110.
  • the paper discharge unit 160 includes a paper discharge stand 162.
  • the paper discharge tray 162 stacks and collects the sheets P conveyed by the conveyance belt 116.
  • the paper output tray 162 is provided with a front sheet contact, a rear sheet contact, and a horizontal sheet contact (not shown), and the sheets P are stacked in order.
  • the paper discharge tray 162 is provided so as to be able to move up and down by a lifting device (not shown). Driving of the lifting and lowering device is controlled in conjunction with increase and decrease of the sheets P stacked on the sheet discharge tray 162. As a result, among the sheets P stacked on the sheet discharge tray 162, the top sheet P is always at a constant height.
  • FIG. 10 is a block diagram showing the internal configuration of the inkjet printing apparatus 100.
  • the inkjet printing apparatus 100 includes a user interface 170, a storage unit 172, in addition to the image defect detection device 12, the transport unit 110, the printing unit 120, the imaging unit 130, the drying unit 140, the sorting unit 150, and the paper discharge unit 160.
  • the general control unit 174, the conveyance control unit 176, the print control unit 178, the imaging control unit 180, the drying control unit 182, the sorting control unit 184, and the paper discharge control unit 186 are provided.
  • the user interface 170 includes an input unit (not shown) and a display unit (not shown) for the user to operate the inkjet printing apparatus 100.
  • the input unit is, for example, an operation panel that receives an input from a user.
  • the display unit is, for example, a display that displays image data and various information. The user can operate the user interface 170 to cause the inkjet printing apparatus 100 to print a desired image.
  • the storage unit 172 stores a program for controlling the inkjet printing apparatus 100 and information necessary for executing the program.
  • the storage unit 172 is configured by a non-temporary recording medium such as a hard disk (not shown) or various semiconductor memories.
  • the storage unit 172 may include volatile memory such as a random access memory (RAM) (not shown) that temporarily stores the reference image D1 and the like.
  • RAM random access memory
  • the overall control unit 174 performs various processes in accordance with the program stored in the storage unit 172, and generally controls the overall operation of the inkjet printing apparatus 100.
  • the conveyance control unit 176 causes the conveyance unit 110 to convey the sheet P in the conveyance direction by controlling a motor (not shown) of the conveyance unit 110.
  • the sheet P supplied from the sheet feeding unit passes through the position facing the printing unit 120, the imaging unit 130, the drying unit 140, and the sorting unit 150, and is finally discharged to the paper delivery unit 160. Be done.
  • the print control unit 178 controls the ejection of the ink by the inkjet heads 122C, 122M, 122Y, and 122K.
  • the print control unit 178 applies ink droplets of cyan, magenta, yellow, and black onto the sheet P at timing when the sheet P passes the positions facing the respective nozzle surfaces by the inkjet heads 122C, 122M, 122Y, and 122K. Direct the discharge. Thereby, a color image is formed on the printing surface of the sheet P, and the sheet P becomes a "printed material".
  • the imaging control unit 180 causes the imaging unit 130 to read the image of the sheet P (printed matter) by controlling the imaging by the scanner 132.
  • the imaging control unit 180 causes the scanner 132 to read the image formed on the sheet P at the timing when the sheet P passes the position facing the scanner 132. Thereby, an examination image is acquired.
  • the drying control unit 182 controls the heating by the heater 142 to dry the sheet P by the drying unit 140.
  • the drying control unit 182 causes the heater 142 to heat the sheet P when the sheet P passes a position facing the heater 142.
  • the sorting control unit 184 causes the sorting unit 150 to sort the sheets P by controlling the stamp processing by the stamper 152.
  • the selection control unit 184 performs the stamp processing by the stamper 152.
  • the discharge control unit 186 controls stacking of the sheets P by the discharge tray 162.
  • the sheet P is discharged to the discharge tray 162 and stacked. Ink adheres to the leading edge of the sheet P of the defective product printed matter. For this reason, the user can specify a defective product print out of the sheets P stacked on the sheet discharge tray 162.
  • FIG. 11 is a flowchart showing processing of the printing method and data used for each processing.
  • the parts common to the flowchart shown in FIG. 4 are denoted by the same reference numerals, and the detailed description thereof is omitted.
  • the inkjet printing apparatus 100 prints an image of one page on one sheet of paper P.
  • step S41 an example of an image acquisition step
  • the image acquisition unit 20 an example of a reference image acquisition unit of the image defect detection device 12 acquires a reference image D1 from the user interface 170 or the storage unit 172.
  • reference images D1 of all pages are acquired.
  • step S42 conveyance of the sheet P is started in the conveyance unit 110.
  • the printing unit 120 prints on the sheet P based on the print source data acquired from the user interface 170 or the storage unit 172.
  • step S43 an example of the image acquisition process
  • the image of the sheet P is imaged by the imaging unit 130 to acquire the inspection image D2.
  • step S2 the first detection processing unit 22 of the image defect detection device 12 performs a first detection process to detect the defect candidate position D3 of the inspection image D2 and the intensity thereof.
  • step S11 the selection unit 24 of the image defect detection device 12 selects the number of selected defect candidate positions D5 equal to or less than the upper limit value D4 from among the plurality of defect candidate positions D3.
  • the selection unit 24 preferentially selects the defect candidate position D3 whose strength is weak.
  • the selection unit 24 sets a defect candidate position other than the selected defect candidate position D5 among the plurality of defect candidate positions D3 as a non-selected defect candidate position D11.
  • the non-selected defect candidate position D11 is input to the determination unit 28 of the image defect detection device 12.
  • the determination unit 28 determines that an image defect exists at the input non-selected defect candidate position D11.
  • step S4 the second detection processing unit 26 of the image defect detection device 12 detects the image defect D6 at the selected defect candidate position D5.
  • step S13 the quality determination unit 30 of the image defect detection device 12 determines the printed matter based on the image defect D6 according to the determination result D12 of the determination unit 28 and the image defect D6 which is the detection result of the second detection processing unit 26. A pass / fail judgment is made, and a pass / fail judgment result D13 is output. The quality determination result D13 is input to the selection control unit 184.
  • step S44 the selection control unit 184 determines whether or not the sheet P passing the position facing the stamper 152 is a non-defective printed material based on the quality determination result D13. If it is not a non-defective print, that is, if it is a non-defective print, the process proceeds to step S45. If the print is non-defective, the process proceeds to step S46.
  • step S45 the stamp processing by the stamper 152 is performed. This facilitates identification of defective printed matter. After performing the stamp processing, the process proceeds to step S46.
  • step S46 the general control unit 174 determines whether printing of all pages of the print job is completed. If there is a page for which printing has not been completed, the process proceeds to S42, and printing of the next sheet P is performed.
  • the image defect detection process When the image defect detection process is performed online as in the inkjet printing apparatus 100, the image defect detection process needs to be completed within the print time of one page (within the print time per sheet of paper P).
  • the defect for performing the second detection process according to the printing time for one page, which is the target time of the detection time, and the processing capability of the first detection processing unit 22 and the second detection processing unit 26. Since the upper limit of the number of candidate positions can be set, it does not occur that the image defect detection process does not fall within the target time. Therefore, it is possible to complete the process within the target time while properly performing the image defect detection process.
  • the inkjet printing apparatus to which the image defect detection apparatus 12 is applied has been described, but the same applies to the case where the image defect detection apparatuses 14 and 16 are applied.
  • the general control unit 174 or the like may determine the quality of the printed matter based on the detected image defect D6.
  • printing is started after acquiring the reference image D1, but the reference image D1 does not necessarily have to be acquired before the start of printing.
  • the captured image of the first page may be set as the reference image D1
  • the image defect detection process may be started from the second page. That is, the reference image D1 may be acquired before the timing at which the first detection process is performed.
  • each process may be performed in parallel, such as performing inspection image reading of the next page while performing image defect processing of a certain page, and further performing printing of the next page.
  • page is used, but this is not a term that is limited to a printing apparatus that prints the sheet of paper P. Even in the case of a printing apparatus that prints on roll paper, it is possible to print by dividing into pages and inspect each of them.
  • the image defect detection processing as in the present invention can use a parallel computing unit as described above.
  • the number of defect candidate positions that can be processed simultaneously is determined by the number of operation cores, the processing capacity per single operation core, and the total memory amount. For this reason, it is easy to set the upper limit value in accordance with the processing capacity of the GPU. Therefore, it is possible to guarantee the end of the image defect detection process within the target time with little effort.
  • the image defect detection method described above is configured as a program for causing a computer to realize each process, and to configure a non-temporary recording medium such as a CD-ROM (Compact Disk-Read Only Memory) storing the program. Is also possible.
  • the hardware-like structure of the processing unit that executes various processes of the image defect detection devices 10, 12, 14, and 16 is various as shown below. It is a processor.
  • Various processors include a CPU (Central Processing Unit), which is a general-purpose processor that executes software (programs) and functions as various processing units, and a GPU (Graphics Processing Unit), which is a processor specialized for image processing, Dedicated to execute specific processing such as Programmable Logic Device (PLD) or ASIC (Application Specific Integrated Circuit), which is a processor whose circuit configuration can be changed after manufacturing an FPGA (Field Programmable Gate Array) or the like It includes a dedicated electric circuit or the like which is a processor having a designed circuit configuration.
  • PLD Programmable Logic Device
  • ASIC Application Specific Integrated Circuit
  • One processing unit may be configured of one of these various types of processors, or two or more processors of the same or different types (for example, a plurality of FPGAs, a combination of a CPU and an FPGA, or a CPU and (A combination of GPUs).
  • a plurality of processing units may be configured by one processor.
  • a plurality of processing units are configured by one processor, first, one processor or more is configured by a combination of one or more CPUs and software as represented by computers such as servers and clients.
  • a processor functions as a plurality of processing units.
  • SoC system on chip
  • IC integrated circuit
  • the hardware-like structure of these various processors is more specifically an electric circuit (circuitry) combining circuit elements such as semiconductor elements.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Image Analysis (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
  • Image Processing (AREA)

Abstract

The present invention provides a printing apparatus, and a device, a method, and a program for detecting image defects, in which image defect detection processing is appropriately performed and finished within a target time. An inspection image obtained by capturing a printout and a reference image serving as a reference for detecting image defects in the printout are acquired. Defective candidate positions indicating candidate positions of the image defects are detected from the inspection image by means of first detection processing. Defective candidate positions, numbering no more than an upper limit value, are selected from among the detected defective candidate positions, and detection of the image defects is only performed at the selected defective candidate positions by means of second detection processing.

Description

印刷装置、画像欠陥検出装置、方法及びプログラムPrinting apparatus, image defect detection apparatus, method and program
　本発明は印刷装置、画像欠陥検出装置、方法及びプログラムに係り、特に印刷物から画像欠陥を検出する技術に関する。 The present invention relates to a printing apparatus, an image defect detection apparatus, a method and a program, and more particularly to a technology for detecting an image defect from a printed matter.
　印刷物は、印刷の際にインク抜け、インク垂れ、キズ、スジ等の画像欠陥が発生する場合がある。このため、印刷物の画像欠陥を検出する自動検査が行われている。画像欠陥を検出する方法として、撮像装置で印刷物を撮像して検査画像を取得し、予め取得した基準画像と検査画像とを画像処理して比較する方法が知られている。この画像処理には、あまり複雑でない基本的な画像処理が使用されることが多い。例えば、基準画像と検査画像とに何らかの基本的な前処理を施した後に両画像の差分画像を求め、差分画像に対して閾値処置を施すことで閾値を超えた信号を欠陥として検出する。前処理としては両画像の位置合わせを行ったり、ノイズ成分の削減のための空間フィルタ処理（ぼかし処理等）を行ったり、あるいは特定の欠陥を強調するための空間フィルタ処理（エッジ強調等）を行ったりする。 Printed matter may cause image defects such as ink loss, ink dripping, scratches, streaks and the like during printing. For this reason, automatic inspection for detecting an image defect of a printed matter is performed. As a method of detecting an image defect, there is known a method of imaging a printed matter with an imaging device to acquire an inspection image, and image processing and comparing a reference image acquired in advance and the inspection image. This image processing often uses less complex basic image processing. For example, after performing some basic preprocessing on the reference image and the inspection image, a differential image of the two images is obtained, and threshold processing is performed on the differential image to detect a signal exceeding the threshold as a defect. As preprocessing, alignment of both images, spatial filter processing (blurring processing etc.) for reduction of noise components, or spatial filter processing (edge emphasis etc) for emphasizing a specific defect To go.
　特許文献１には、印刷物の全面の詳細な検査を高速に行う技術が記載されている。 Patent Document 1 describes a technique for performing a detailed inspection of the entire surface of a printed matter at high speed.
特開２０００－１７２８４４号公報Unexamined-Japanese-Patent No. 2000-172844
　画像欠陥には様々な「レベル」が存在する。例えばラインヘッド型インクジェット印刷装置では、インクジェットヘッドのノズルに吐出曲がり又は不吐出等のノズル不良が発生すると、そのノズルの印字位置にスジ状の欠陥が発生する。このスジ欠陥は、ノズルが不吐出の場合にはそのノズルの印字位置に全く印字されないため、非常に強いレベルとなる。これに対し、ノズルが吐出曲がりの場合には、曲がり量が大きい場合は強いレベルとなり、曲がり量が小さい場合には弱いレベルとなる。 There are various "levels" of image defects. For example, in a line head type inkjet printing apparatus, when a nozzle defect such as ejection bending or non-ejection occurs in a nozzle of an inkjet head, a stripe defect occurs in the printing position of the nozzle. This streak defect is a very strong level because it is not printed at all at the printing position of the nozzle when the nozzle is not discharged. On the other hand, when the nozzle has a discharge curve, the level is strong when the amount of curve is large, and the level is weak when the amount of curve is small.
　画像欠陥の検出において、製品品質としては問題のない弱いレベルの画像欠陥を過剰に検出してしまうと、印刷物としての歩留まりが悪くなる。一方で、強いレベル、あるいは中程度のレベルの画像欠陥を見逃してしまうと、不良品印刷物を出荷することになる。したがって、画像欠陥のレベルに応じて検出するか否かを判断することが望ましい。 In the detection of the image defect, if a weak level image defect which does not have any problem as a product quality is excessively detected, the yield as a printed matter is deteriorated. On the other hand, if a strong or medium level image defect is missed, a defective printed material will be shipped. Therefore, it is desirable to determine whether to detect in accordance with the level of the image defect.
　画像欠陥のレベルに応じた検出の有無を実現するために、閾値等のパラメータ調整を行うことが多い。パラメータ調整は、メーカー側で行われることもあれば、ＵＩ（User Interface）を介してクライアント側に行わせる場合もある。 In order to realize the presence or absence of detection according to the level of the image defect, parameter adjustment such as a threshold is often performed. The parameter adjustment may be performed by the manufacturer or may be performed by the client via a UI (User Interface).
　画像欠陥のレベルは、最終的には人の目視判断で決まる。しかしながら、人の視覚系と撮像装置では特性が異なるため、基本的な画像処理に対して単純にパラメータ調整するだけでは十分に対応することができない。例えば、人の視覚系と撮像装置では色階調特性が異なるため、ある色階調では弱いレベルと強いレベルの検出有無を適切に設定できた場合であっても、他の階調では適切に設定することができない、ということが発生する。また、同じ欠陥であっても、視覚系では欠陥周辺の背景が平網（空間周波数が低い）に近いほど強いレベルと認識し、背景が複雑（空間周波数が高い）であるほど弱いレベルと認識する。 The level of image defects is ultimately determined by human visual judgment. However, since characteristics of the human visual system and the image pickup apparatus are different, it is not possible to sufficiently cope with basic image processing simply by parameter adjustment. For example, since color tone characteristics are different between the human visual system and the imaging device, even if it is possible to appropriately set the presence or absence of a weak level and a strong level in a certain color tone, it is appropriate in another tone. It happens that it can not be set. Also, even with the same defect, in the visual system, the background around the defect is recognized as a stronger level as it is closer to the flat network (the spatial frequency is lower), and as the background is more complex (the spatial frequency is higher) as a weaker level Do.
　しかしながら、人の視覚系を模した複雑な欠陥検出をしようとすると複雑な画像処理演算が必要となる。このため、演算時間が目標時間に対して長すぎてしまい、導入することができないといった問題が発生する。 However, complex image processing operations are required to detect complex defects that mimic human visual systems. For this reason, the operation time is too long for the target time, which causes a problem that it can not be introduced.
　また、特許文献１に記載の技術においても、目標時間内に検査が完了しない可能性があった。 Further, even with the technique described in Patent Document 1, there is a possibility that the inspection may not be completed within the target time.
　本発明はこのような事情に鑑みてなされたもので、画像欠陥検出処理を適切に行いつつ、目標時間内に終了させる印刷装置、画像欠陥検出装置、方法及びプログラムを提供することを目的とする。 The present invention has been made in view of such circumstances, and it is an object of the present invention to provide a printing apparatus, an image defect detection apparatus, a method, and a program which are terminated within a target time while appropriately performing image defect detection processing. .
　上記目的を達成するために画像欠陥検出装置の一の態様は、印刷物を撮像した検査画像と印刷物の画像欠陥を検出する基準となる基準画像とを取得する画像取得部と、第１検出処理により、検査画像から画像欠陥の候補の位置を示す欠陥候補位置を検出する第１検出処理部と、検査画像と基準画像とを比較する第２検出処理により、欠陥候補位置の画像欠陥を検出する第２検出処理部と、検出された欠陥候補位置のうち第２検出処理を行う欠陥候補位置を選択する選択部と、を備え、選択部は、上限値以下の数の欠陥候補位置を選択し、第２検出処理部は、選択された欠陥候補位置についてのみ第２検出処理を行う画像欠陥検出装置である。 In order to achieve the above object, one aspect of the image defect detection apparatus includes an image acquisition unit for acquiring an inspection image obtained by imaging a printed matter and a reference image serving as a reference for detecting an image defect of the printed matter; A first detection processing unit that detects a defect candidate position indicating a position of an image defect candidate from an inspection image, and a second detection process that compares the inspection image with a reference image to detect an image defect at the defect candidate position And (2) a detection processing unit, and a selection unit that selects a defect candidate position to be subjected to the second detection process among the detected defect candidate positions, the selection unit selects a number of defect candidate positions equal to or less than the upper limit value, The second detection processing unit is an image defect detection device that performs the second detection process only for the selected defect candidate position.
　本態様によれば、印刷物を撮像した検査画像と印刷物の画像欠陥を検出する基準となる基準画像とを取得し、第１検出処理により検査画像から画像欠陥の候補の位置を示す欠陥候補位置を検出し、検出された欠陥候補位置のうち第２検出処理を行う欠陥候補位置であって上限値以下の数の欠陥候補位置を選択し、選択された欠陥候補位置についてのみ検査画像と基準画像とを比較する第２検出処理により欠陥候補位置の画像欠陥を検出するようにしたので、画像欠陥検出処理を適切に行いつつ、目標時間内に終了させることができる。 According to this aspect, the inspection image obtained by imaging the printed matter and the reference image serving as the reference for detecting the image defect of the printed matter are acquired, and the defect detection positions indicating the positions of the image defect candidates from the inspected image are obtained by the first detection process. Among the detected defect candidate positions, the defect candidate positions to be subjected to the second detection process and having a number equal to or less than the upper limit are selected, and the inspection image and the reference image are selected only for the selected defect candidate positions. Since the image defect of the defect candidate position is detected by the second detection process of comparing the above, it is possible to finish within the target time while appropriately performing the image defect detection process.
　第１検出処理部は、欠陥候補位置の強度を検出し、選択部は、検出した強度が弱い欠陥候補位置を優先して選択し、欠陥候補位置のうち選択部において選択されない欠陥候補位置については画像欠陥が存在すると判断する判断部を備えることが好ましい。これにより、検出した強度に応じて欠陥候補位置を選択し、選択されない欠陥候補位置について適切に判断することができる。 The first detection processing unit detects the strength of the defect candidate position, the selection unit preferentially selects the defect candidate position where the detected strength is weak, and the defect candidate position not selected by the selection unit among the defect candidate positions It is preferable to include a determination unit that determines that an image defect exists. Thereby, the defect candidate position can be selected according to the detected intensity, and the defect candidate position not selected can be appropriately determined.
　第１検出処理部は、欠陥候補位置の強度を検出し、選択部は、検出した強度が強い欠陥候補位置を優先して選択することが好ましい。これにより、検出した強度に応じて欠陥候補位置を選択することができる。 Preferably, the first detection processing unit detects the strength of the defect candidate position, and the selection unit preferentially selects a defect candidate position having a high detected strength. Thereby, the defect candidate position can be selected according to the detected intensity.
　欠陥候補位置のうち選択部において選択されない欠陥候補位置については画像欠陥が存在しないと判断する判断部を備えることが好ましい。これにより、選択されない欠陥候補位置について適切に判断することができる。 It is preferable to include a determination unit that determines that an image defect does not exist at a defect candidate position that is not selected by the selection unit among the defect candidate positions. Thereby, it is possible to appropriately determine the defect candidate position which is not selected.
　欠陥候補位置を検出した順に格納する格納部を備え、選択部は、欠陥候補位置を格納部に格納された正順又は逆順に選択することが好ましい。これにより、欠陥候補位置の強度によるソート処理等の演算リソースを別途使用することがないので、選択部の上限値を大きくすることができる。 It is preferable that the storage unit stores the defect candidate positions in the order of detection, and the selection unit selects the defect candidate positions in the normal order or the reverse order stored in the storage unit. As a result, there is no need to separately use computation resources such as sorting processing based on the strength of the defect candidate position, so the upper limit value of the selection unit can be increased.
　欠陥候補位置の数が上限値を超えた場合にユーザに通知する通知部を備えることが好ましい。これにより、ユーザは上限値を超えたことを知ることができる。 It is preferable to provide a notification unit that notifies the user when the number of defect candidate positions exceeds the upper limit value. This allows the user to know that the upper limit has been exceeded.
　通知部による通知の有無を設定する設定部を備えることが好ましい。これにより、ユーザは必要な場合のみ上限値を超えたことを知ることができる。 Preferably, a setting unit configured to set presence / absence of notification by the notification unit is provided. This allows the user to know that the upper limit has been exceeded only when necessary.
　画像全体に対する第２検出処理の演算量は、画像全体に対する第１検出処理の演算量よりも大きいことが好ましい。また、画像の単一領域に対する第２検出処理の演算量は、画像の単一領域に対する第１検出処理の演算量よりも大きいことが好ましい。第２検出処理を演算量の大きい演算とすることで、高精度な検出処理を行うことができる。 It is preferable that the operation amount of the second detection process for the entire image is larger than the operation amount of the first detection process for the entire image. Further, it is preferable that the operation amount of the second detection process for a single area of the image is larger than the operation amount of the first detection process for a single area of the image. By making the second detection process a calculation with a large amount of calculation, highly accurate detection process can be performed.
　第１検出処理は、検査画像と基準画像とを使用することが好ましい。これにより、欠陥候補位置を適切に検出することができる。 Preferably, the first detection process uses an inspection image and a reference image. Thereby, the defect candidate position can be appropriately detected.
　第２検出処理は、欠陥候補位置を含む検査部分画像と検査部分画像に対応する基準部分画像を使用することが好ましい。これにより、欠陥候補位置の画像欠陥を適切に検出することができる。 The second detection process preferably uses an inspection partial image including a defect candidate position and a reference partial image corresponding to the inspection partial image. Thereby, the image defect of the defect candidate position can be appropriately detected.
　第２検出処理は周波数解析処理を含むことが好ましい。これにより、人の視覚系に対応させた検出を行うことができる。 The second detection process preferably includes a frequency analysis process. Thereby, detection corresponding to the human visual system can be performed.
　第２検出処理は多重解像度解析処理を含むことが好ましい。これにより、人の視覚系に対応させた検出を行うことができる。 Preferably, the second detection process comprises a multiresolution analysis process. Thereby, detection corresponding to the human visual system can be performed.
　第２検出処理は機械学習モデル又は深層学習モデルによる判定を含むことが好ましい。これにより、人の視覚系に対応させた検出を行うことができる。 The second detection process preferably includes determination by a machine learning model or a deep learning model. Thereby, detection corresponding to the human visual system can be performed.
　機械学習モデル又は深層学習モデルを作成するための学習データには、人の官能評価実験によって得られた官能値又は官能値を加工したデータを含むことが好ましい。これにより、機械学習モデル又は深層学習モデルが人の視覚特性を表現することができ、人の視覚系に対応させた検出を行うことができる。 The learning data for creating a machine learning model or a deep learning model preferably includes data obtained by processing sensory values or sensory values obtained by human sensory evaluation experiments. Thereby, the machine learning model or the deep learning model can express human visual characteristics, and detection corresponding to the human visual system can be performed.
　第２検出処理部は、複数の演算を並列に処理する並列演算器を備えることが好ましい。これにより、第２検出処理を高速に行うことができ、上限値を大きくすることができる。 The second detection processing unit preferably includes a parallel computing unit that processes a plurality of operations in parallel. Thereby, the second detection process can be performed at high speed, and the upper limit value can be increased.
　並列演算器は複数のＣＰＵ（Central Processing Unit）を備えることが好ましい。また、並列演算器は複数の基板を備えることが好ましい。さらに、並列演算器は複数のＧＰＵ（Graphics Processing Unit）を備えてもよい。このように構成することで、並列演算器を適切に構成することができる。 Preferably, the parallel computing unit comprises a plurality of CPUs (Central Processing Units). Preferably, the parallel computing unit comprises a plurality of substrates. Furthermore, the parallel computing unit may include a plurality of GPUs (Graphics Processing Units). By configuring in this manner, the parallel computing unit can be appropriately configured.
　上記目的を達成するために印刷装置の一の態様は、印刷物を印刷する印刷部と、印刷物を撮像して検査画像を取得する撮像部と、印刷物の画像欠陥を検出する基準となる基準画像を取得する基準画像取得部と、第１検出処理により、検査画像から画像欠陥の候補の位置を示す欠陥候補位置を検出する第１検出処理部と、検査画像と基準画像とを比較する第２検出処理により、欠陥候補位置の画像欠陥を検出する第２検出処理部と、検出された欠陥候補位置のうち第２検出処理を行う欠陥候補位置を選択する選択部と、第２検出処理の検出結果に基づいて検査画像の良否の判定を行う良否判定部と、を備え、選択部は、上限値以下の数の欠陥候補位置を選択し、第２検出処理部は、選択された欠陥候補位置についてのみ第２検出処理を行う印刷装置である。 In order to achieve the above object, one aspect of the printing apparatus is a printing unit for printing a printed matter, an imaging unit for imaging a printed matter to acquire an inspection image, and a reference image serving as a reference for detecting an image defect of the printed matter A reference image acquisition unit to acquire, a first detection processing unit that detects a defect candidate position indicating a position of an image defect candidate from an inspection image by a first detection process, and a second detection that compares the inspection image and the reference image A second detection processing unit that detects an image defect at a defect candidate position by processing, a selection unit that selects a defect candidate position at which the second detection processing is performed among the detected defect candidate positions, and a detection result of the second detection processing And the selection unit selects defect candidate positions having a number equal to or less than the upper limit value, and the second detection processing unit determines the selected defect candidate positions. Only the second detection processing mark It is a device.
　本態様によれば、印刷物を印刷し、印刷物を撮像して検査画像を取得し、印刷物の画像欠陥を検出する基準となる基準画像を取得し、第１検出処理により検査画像から画像欠陥の候補の位置を示す欠陥候補位置を検出し、検出された欠陥候補位置のうち第２検出処理を行う欠陥候補位置であって上限値以下の数の欠陥候補位置を選択し、選択された欠陥候補位置についてのみ検査画像と基準画像とを比較する第２検出処理により欠陥候補位置の画像欠陥を検出し、第２検出処理の検出結果に基づいて検査画像の良否の判定を行うようにしたので、画像欠陥検出処理を適切に行いつつ、検査画像の良否判定を目標時間内に終了させることができる。 According to this aspect, the printed matter is printed, the printed matter is imaged, the inspection image is acquired, the reference image serving as the reference for detecting the image defect of the printed matter is acquired, and the candidate for the image defect from the inspection image is obtained by the first detection process. Detects the defect candidate position indicating the position of the defect candidate position, selects the number of defect candidate positions that are the defect candidate positions to be subjected to the second detection processing among the detected defect candidate positions and not more than the upper limit, and selects the selected defect candidate positions Since the image defect of the defect candidate position is detected by the second detection process of comparing the inspection image and the reference image only for the above, the quality of the inspection image is judged based on the detection result of the second detection process. The quality determination of the inspection image can be completed within the target time while appropriately performing the defect detection processing.
　上記目的を達成するために画像欠陥検出方法の一の態様は、印刷物を撮像した検査画像と印刷物の画像欠陥を検出する基準となる基準画像とを取得する画像取得工程と、第１検出処理により、検査画像から画像欠陥の候補の位置を示す欠陥候補位置を検出する第１検出処理工程と、検査画像と基準画像とを比較する第２検出処理により、欠陥候補位置の画像欠陥を検出する第２検出処理工程と、検出された欠陥候補位置のうち第２検出処理を行う欠陥候補位置を選択する選択工程と、を備え、選択工程は、上限値以下の数の欠陥候補位置を選択し、第２検出処理工程は、選択された欠陥候補位置についてのみ第２検出処理を行う画像欠陥検出方法である。 In order to achieve the above object, one aspect of the image defect detection method includes an image acquisition step of acquiring an inspection image obtained by imaging a printed matter and a reference image serving as a reference for detecting an image defect of the printed matter; A first detection processing step of detecting a defect candidate position indicating a position of a candidate of an image defect from an inspection image, and a second detection process of comparing the inspection image and a reference image detect an image defect of the defect candidate position (2) including a detection process step and a selection step of selecting a defect candidate position to be subjected to the second detection process among the detected defect candidate positions, the selection step selecting defect candidate positions having a number equal to or less than the upper limit value; The second detection process is an image defect detection method in which the second detection process is performed only for the selected defect candidate position.
　本態様によれば、印刷物を撮像した検査画像と印刷物の画像欠陥を検出する基準となる基準画像とを取得し、第１検出処理により検査画像から画像欠陥の候補の位置を示す欠陥候補位置を検出し、検出された欠陥候補位置のうち第２検出処理を行う欠陥候補位置であって上限値以下の数の欠陥候補位置を選択し、選択された欠陥候補位置についてのみ検査画像と基準画像とを比較する第２検出処理により欠陥候補位置の画像欠陥を検出するようにしたので、画像欠陥検出処理を適切に行いつつ、目標時間内に終了させることができる。 According to this aspect, the inspection image obtained by imaging the printed matter and the reference image serving as the reference for detecting the image defect of the printed matter are acquired, and the defect detection positions indicating the positions of the image defect candidates from the inspected image are obtained by the first detection process. Among the detected defect candidate positions, the defect candidate positions to be subjected to the second detection process and having a number equal to or less than the upper limit are selected, and the inspection image and the reference image are selected only for the selected defect candidate positions. Since the image defect of the defect candidate position is detected by the second detection process of comparing the above, it is possible to finish within the target time while appropriately performing the image defect detection process.
　上記目的を達成するために画像欠陥検出方法をコンピュータに実行させるためのプログラムの一の態様は、印刷物を撮像した検査画像と印刷物の画像欠陥を検出する基準となる基準画像とを取得する画像取得工程と、第１検出処理により、検査画像から画像欠陥の候補の位置を示す欠陥候補位置を検出する第１検出処理工程と、検査画像と基準画像とを比較する第２検出処理により、欠陥候補位置の画像欠陥を検出する第２検出処理工程と、検出された欠陥候補位置のうち第２検出処理を行う欠陥候補位置を選択する選択工程と、を備え、選択工程は、上限値以下の数の欠陥候補位置を選択し、第２検出処理工程は、選択された欠陥候補位置についてのみ第２検出処理を行う画像欠陥検出方法をコンピュータに実行させるためのプログラムである。 One aspect of a program for causing a computer to execute the image defect detection method to achieve the above object acquires an inspection image obtained by imaging a printed matter and a reference image serving as a reference for detecting an image defect of the printed matter A defect detection candidate detecting a defect candidate position indicating a position of an image defect candidate from the inspection image by the first detection processing, and a second detection processing comparing the inspection image with the reference image; A second detection process step of detecting an image defect at a position; and a selection step of selecting a defect candidate position at which the second detection process is to be performed among the detected defect candidate positions; Program for causing a computer to execute an image defect detection method in which a second detection processing step performs a second detection process only for the selected defect candidate position. A.
　本態様によれば、印刷物を撮像した検査画像と印刷物の画像欠陥を検出する基準となる基準画像とを取得し、第１検出処理により検査画像から画像欠陥の候補の位置を示す欠陥候補位置を検出し、検出された欠陥候補位置のうち第２検出処理を行う欠陥候補位置であって上限値以下の数の欠陥候補位置を選択し、選択された欠陥候補位置についてのみ検査画像と基準画像とを比較する第２検出処理により欠陥候補位置の画像欠陥を検出するようにしたので、画像欠陥検出処理を適切に行いつつ、目標時間内に終了させることができる。 According to this aspect, the inspection image obtained by imaging the printed matter and the reference image serving as the reference for detecting the image defect of the printed matter are acquired, and the defect detection positions indicating the positions of the image defect candidates from the inspected image are obtained by the first detection process. Among the detected defect candidate positions, the defect candidate positions to be subjected to the second detection process and having a number equal to or less than the upper limit are selected, and the inspection image and the reference image are selected only for the selected defect candidate positions. Since the image defect of the defect candidate position is detected by the second detection process of comparing the above, it is possible to finish within the target time while appropriately performing the image defect detection process.
　上記目的を達成するために画像欠陥検出装置の一の態様は、プロセッサを有する画像欠陥検出装置であって、プロセッサが、印刷物を撮像した検査画像と印刷物の画像欠陥を検出する基準となる基準画像とを取得し、第１検出処理により、検査画像から画像欠陥の候補の位置を示す欠陥候補位置を検出し、検出された欠陥候補位置のうち第２検出処理を行う上限値以下の数の欠陥候補位置を選択し、検査画像と基準画像とを比較する第２検出処理により、選択された欠陥候補位置についてのみ欠陥候補位置の画像欠陥を検出する画像欠陥検出装置である。 In order to achieve the above object, one aspect of the image defect detection apparatus is an image defect detection apparatus having a processor, wherein the processor is a reference image that serves as a reference for detecting an inspection image obtained by imaging a printed material and an image defect of the printed material , And detects a defect candidate position indicating the position of an image defect candidate from the inspection image by the first detection process, and detects the number of defects equal to or less than the upper limit value for performing the second detection process among the detected defect candidate positions. It is an image defect detection device which detects an image defect of a defect candidate position only for the selected defect candidate position by a second detection process of selecting a candidate position and comparing an inspection image and a reference image.
　本発明によれば、画像欠陥検出処理を適切に行いつつ、目標時間内に終了させることができる。 According to the present invention, it is possible to finish within the target time while appropriately performing the image defect detection process.
画像欠陥検出装置の内部構成を示すブロック図Block diagram showing the internal configuration of the image defect detection apparatus 画像欠陥検出方法の処理を示したフローチャートFlow chart showing processing of image defect detection method 画像欠陥検出装置の内部構成を示すブロック図Block diagram showing the internal configuration of the image defect detection apparatus 画像欠陥検出方法の処理を示したフローチャートFlow chart showing processing of image defect detection method 画像欠陥検出方法の処理を示したフローチャートFlow chart showing processing of image defect detection method 画像欠陥検出装置の内部構成を示すブロック図Block diagram showing the internal configuration of the image defect detection apparatus 画像欠陥検出装置の内部構成を示すブロック図Block diagram showing the internal configuration of the image defect detection apparatus 画像欠陥検出方法の処理を示したフローチャートFlow chart showing processing of image defect detection method インクジェット印刷装置の全体構成図Overall configuration of inkjet printing device インクジェット印刷装置の内部構成を示すブロック図Block diagram showing the internal configuration of the ink jet printing apparatus 印刷方法の処理と各処理に用いるデータについて示したフローチャートFlow chart showing processing of printing method and data used for each processing
　以下、添付図面に従って本発明の好ましい実施形態について詳説する。 Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings.
　＜第１の実施形態＞
　〔画像欠陥検出装置の構成〕
　図１は、第１の実施形態に係る画像欠陥検出装置１０の内部構成を示すブロック図である。画像欠陥検出装置１０は、印刷物の画像欠陥を検出する装置である。図１に示すように、画像欠陥検出装置１０は、画像取得部２０、第１検出処理部２２、選択部２４、及び第２検出処理部２６を備えている。 First Embodiment
[Configuration of Image Defect Detection Device]
FIG. 1 is a block diagram showing an internal configuration of the image defect detection apparatus 10 according to the first embodiment. The image defect detection device 10 is a device that detects an image defect of a printed matter. As shown in FIG. 1, the image defect detection apparatus 10 includes an image acquisition unit 20, a first detection processing unit 22, a selection unit 24, and a second detection processing unit 26.
　画像取得部２０は、画像欠陥を検出する印刷物を撮像した検査画像を取得する。また、画像取得部２０は、印刷物の画像欠陥を検出する基準となる基準画像を取得する。画像取得部２０は、検査画像及び基準画像を、装置に備えられた不図示の撮像装置から取得してもよいし、不図示の入力インターフェースから取得してもよい。 The image acquisition unit 20 acquires an inspection image obtained by imaging a printed matter for detecting an image defect. Further, the image acquisition unit 20 acquires a reference image which is a reference for detecting an image defect of a printed matter. The image acquisition unit 20 may acquire an inspection image and a reference image from an imaging device (not shown) provided in the device, or may acquire it from an input interface (not shown).
　第１検出処理部２２は、第１検出処理により、検査画像から画像欠陥の候補の位置を示す欠陥候補位置を検出する。第１検出処理の詳細については、後述する。 The first detection processing unit 22 detects a defect candidate position indicating a position of an image defect candidate from the inspection image by the first detection process. Details of the first detection process will be described later.
　選択部２４は、第１検出処理により検出された欠陥候補位置のうち、後述する第２検出処理を行う欠陥候補位置を選択する。選択部２４は、予め定められた上限値以下の範囲の数の欠陥候補位置を選択する。 The selection unit 24 selects, from among the defect candidate positions detected by the first detection process, a defect candidate position to be subjected to a second detection process described later. The selection unit 24 selects the number of defect candidate positions in the range of the predetermined upper limit value or less.
　第２検出処理部２６は、検査画像と基準画像とを比較する第２検出処理により、欠陥候補位置の画像欠陥を検出する。第２検出処理の詳細については後述する。 The second detection processing unit 26 detects an image defect of the defect candidate position by a second detection process of comparing the inspection image and the reference image. Details of the second detection process will be described later.
　〔画像欠陥検出方法〕
　画像欠陥検出装置１０における画像欠陥検出方法（画像欠陥検出装置１０の動作方法）について説明する。図２は、画像欠陥検出方法の処理と各処理に用いるデータについて示したフローチャートである。 [Image defect detection method]
The image defect detection method (the operation method of the image defect detection device 10) in the image defect detection device 10 will be described. FIG. 2 is a flowchart showing processing of the image defect detection method and data used for each processing.
　画像欠陥検出が指示されると、ステップＳ１（画像取得工程の一例）において、画像取得部２０は、画像欠陥を検出する基準となる基準画像Ｄ１と、画像欠陥を検出しようとする印刷物を撮像した検査画像Ｄ２とを取得する。本実施形態における基準画像Ｄ１は、画像欠陥を検出する印刷物と同じ入力画像データ（印刷元データ）に基づいて印刷された良品印刷物を、撮像装置で撮像した画像データである。画像取得部２０は、不図示のメモリに予め記憶された基準画像Ｄ１を読み出して取得する。画像取得部２０は、不図示の入力インターフェースから基準画像Ｄ１を取得してもよい。また、画像取得部２０は、不図示の入力インターフェースから検査画像Ｄ２を取得する。 When an image defect detection is instructed, in step S1 (an example of an image acquisition step), the image acquiring unit 20 captures a reference image D1 serving as a reference for detecting an image defect and a printed matter for detecting an image defect. An inspection image D2 is acquired. The reference image D1 in the present embodiment is image data obtained by imaging the non-defective printed material printed based on the same input image data (print source data) as the printed material for which an image defect is detected, by the imaging device. The image acquisition unit 20 reads out and acquires a reference image D1 stored in advance in a memory (not shown). The image acquisition unit 20 may acquire the reference image D1 from an input interface (not shown). Further, the image acquisition unit 20 acquires the inspection image D2 from an input interface (not shown).
　なお、基準画像Ｄ１は、撮像装置で撮像された画像に限定されない。例えば、印刷物を印刷した際の印刷装置の入力画像データを基準画像として用いてもよい。また、入力画像データに解像度変換処理、色変換処理、階調変換処理、又はスクリーニング処理等を施したものを基準画像Ｄ１として使用してもよい。 The reference image D1 is not limited to the image captured by the imaging device. For example, input image data of the printing apparatus when printing a printed matter may be used as a reference image. Further, input image data subjected to resolution conversion processing, color conversion processing, gradation conversion processing, screening processing or the like may be used as the reference image D1.
　取得された基準画像Ｄ１及び検査画像Ｄ２は、第１検出処理部２２及び第２検出処理部２６に入力される。 The acquired reference image D1 and inspection image D2 are input to the first detection processing unit 22 and the second detection processing unit 26.
　続いて、ステップＳ２（第１検出処理工程の一例）では、第１検出処理部２２において第１検出処理を行い、検査画像Ｄ２の欠陥候補位置Ｄ３を検出する。検出された欠陥候補位置Ｄ３は、選択部２４に入力される。 Subsequently, in step S2 (an example of a first detection process), the first detection processing unit 22 performs the first detection process to detect the defect candidate position D3 of the inspection image D2. The detected defect candidate position D3 is input to the selection unit 24.
　ステップＳ３（選択工程の一例）では、選択部２４は、複数の欠陥候補位置Ｄ３の中から上限値Ｄ４以下の数の選択欠陥候補位置Ｄ５を選択する。上限値Ｄ４は、画像欠陥を検出するための目標時間に応じて予め定められた値である。上限値Ｄ４は、不図示のメモリに予め記憶されている。選択された選択欠陥候補位置Ｄ５は、第２検出処理部２６に入力される。 In step S3 (an example of a selection step), the selection unit 24 selects the number of selected defect candidate positions D5 equal to or less than the upper limit value D4 from among the plurality of defect candidate positions D3. The upper limit value D4 is a value predetermined in accordance with a target time for detecting an image defect. The upper limit value D4 is stored in advance in a memory (not shown). The selected selected defect candidate position D5 is input to the second detection processing unit 26.
　最後にステップＳ４（第２検出処理工程の一例）では、第２検出処理部２６は、基準画像Ｄ１と検査画像Ｄ２とに基づいて第２検出処理を行い、それぞれの選択欠陥候補位置Ｄ５に対して画像欠陥Ｄ６を検出する。 Finally, in step S4 (an example of a second detection processing step), the second detection processing unit 26 performs a second detection processing based on the reference image D1 and the inspection image D2, and applies to each selected defect candidate position D5. Image defect D6 is detected.
　以上で、画像欠陥検出が終了する。 This is the end of the image defect detection.
　〔第１検出処理〕
　第１検出処理は、基準画像Ｄ１と検査画像Ｄ２とを使用する。第１検出処理部２２は、第１検出処理の前処理として、基準画像Ｄ１及び検査画像Ｄ２の解像度合わせ処理、階調合わせ処理、明るさ合わせ処理、色合わせ処理、位置合わせ処理、ノイズ低減のための空間フィルタ処理、画像欠陥を強調するための空間フィルタ処理を実施してもよい。 [First detection processing]
The first detection process uses the reference image D1 and the inspection image D2. The first detection processing unit 22 performs resolution matching processing of the reference image D1 and the inspection image D2, gradation matching processing, brightness matching processing, color matching processing, alignment processing, noise reduction processing as pre-processing of the first detection processing. For the purpose of spatial filtering, spatial filtering for enhancing image defects may be performed.
　第１検出処理は、例えば、基準画像Ｄ１と検査画像Ｄ２の差分画像を生成し、差分画像の各画素の絶対値がある閾値を超えた位置を欠陥候補位置とする処理である。 The first detection process is, for example, a process of generating a difference image between the reference image D1 and the inspection image D2 and setting a position where the absolute value of each pixel of the difference image exceeds a certain threshold as a defect candidate position.
　第１検出処理は、基準画像Ｄ１と検査画像Ｄ２との差分画像（Ｄ２－Ｄ１）の負値を０にクリッピングした上で、各画素の画素値がある閾値を超えた位置を欠陥候補位置とする処理であってもよい。この場合、基準画像Ｄ１より明るい欠陥候補位置を検出することができる。 In the first detection processing, after clipping the negative value of the difference image (D2-D1) between the reference image D1 and the inspection image D2 to 0, the position where the pixel value of each pixel exceeds a threshold is the defect candidate position It may be a process to In this case, a defect candidate position brighter than the reference image D1 can be detected.
　また、第１検出処理は、基準画像Ｄ１と検査画像Ｄ２との差分画像（Ｄ１－Ｄ２）の負値を０にクリッピングした上で、各画素の画素値がある閾値を超えたところを欠陥候補位置とする処理であってもよい。この場合、基準画像Ｄ１より暗い欠陥候補位置を検出することができる。 Further, in the first detection processing, after clipping the negative value of the difference image (D1-D2) between the reference image D1 and the inspection image D2 to 0, the pixel value of each pixel exceeds a certain threshold to be a defect candidate It may be processing to be a position. In this case, a defect candidate position darker than the reference image D1 can be detected.
　また、第１検出処理は、領域毎に行う処理であってもよい。例えば、最初に、基準画像Ｄ１と検査画像Ｄ２とをそれぞれ格子状に分割する。次に、格子毎に差分格子画像を生成する。さらに、この差分格子画像の絶対値の統計値（平均値、中央値等）を算出する。最後に、この統計値とある閾値とを比較する。そして、閾値を超えた場合に、その差分格子画像に対応する格子の中央位置を欠陥候補位置とする。なお、格子への分割は、重複を許して実施してもよい。 The first detection process may be a process performed for each area. For example, first, the reference image D1 and the inspection image D2 are divided into grids, respectively. Next, a difference grid image is generated for each grid. Further, statistical values (average value, median value, etc.) of the absolute values of this difference grid image are calculated. Finally, this statistic is compared with a threshold. When the threshold value is exceeded, the center position of the grid corresponding to the difference grid image is set as the defect candidate position. Note that the division into grids may be performed with overlapping allowed.
　さらに、検出対象の画像欠陥がスジ欠陥のように特定の方向性を有している場合は、第１検出処理は、以下のような処理であってもよい。 Furthermore, in the case where the image defect to be detected has a specific directionality like a streak defect, the first detection process may be the following process.
　まず、基準画像Ｄ１と検査画像Ｄ２とをそれぞれ格子状に分割する。次に、基準画像Ｄ１を格子毎にスジ欠陥の方向に合わせて積算処理した基準画像プロファイルを生成する。同様に、検査画像Ｄ２を格子毎にスジ欠陥の方向に合わせて積算処理した検査画像プロファイルを生成する。続いて、基準画像プロファイルと検査画像プロファイルとの差分プロファイルを生成する。この差分プロファイルの強度値がある閾値を超えた場合に、その差分プロファイルに対応する格子を欠陥候補位置とする。なお、格子への分割は重複を許して実施してもよい。 First, the reference image D1 and the inspection image D2 are divided into grids, respectively. Next, a reference image profile is generated in which the reference image D1 is integrated for each grid in accordance with the direction of the streak defect. Similarly, an inspection image profile is generated in which the inspection image D2 is integrated for each grid in accordance with the direction of the streak defect. Subsequently, a difference profile between the reference image profile and the inspection image profile is generated. When the intensity value of the difference profile exceeds a certain threshold, the lattice corresponding to the difference profile is set as a defect candidate position. Note that the division into lattices may be performed with overlapping allowed.
　第１検出処理として、これらの処理を複数実施してもよい。なお、第１検出処理は、以上の処理に限定されず、検査画像から欠陥候補位置を検出できればよい。 A plurality of these processes may be performed as the first detection process. The first detection process is not limited to the above process, as long as the defect candidate position can be detected from the inspection image.
　〔第２検出処理〕
　第２検出処理は、人の認識に近い画像欠陥を行う。第２検出処理は、例えば以下の処理である。まず、選択欠陥候補位置Ｄ５毎に、基準画像Ｄ１及び検査画像Ｄ２のそれぞれの選択欠陥候補位置Ｄ５を中心とした部分画像である基準部分画像及び検査部分画像を取得する。なお、基準部分画像及び検査部分画像は、選択欠陥候補位置Ｄ５を中心としたものに限定されず、選択欠陥候補位置Ｄ５を含んでいればよい。 [Second detection process]
The second detection process performs an image defect close to human recognition. The second detection process is, for example, the following process. First, for each of the selected defect candidate positions D5, a reference partial image and an inspection partial image which are partial images centering on the selected defect candidate positions D5 of the reference image D1 and the inspection image D2 are acquired. The reference partial image and the inspection partial image are not limited to those centered on the selected defect candidate position D5, and may include the selected defect candidate position D5.
　次に、基準部分画像及び検査部分画像から差分部分画像を生成する。続いて、これら基準部分画像、検査部分画像、及び差分部分画像の少なくとも一部を用いて、その領域の特徴量を算出する。この特徴量を機械学習モデル（重回帰分析、部分的最小二乗法、ニューラルネットワーク、サポートベクターマシン、リッジ回帰等）に入力することで、選択欠陥候補位置Ｄ５の欠陥候補が画像欠陥Ｄ６であるか、又は非欠陥であるかの判定（分類）を行う。 Next, a difference partial image is generated from the reference partial image and the inspection partial image. Subsequently, the feature amount of the area is calculated using at least a part of the reference partial image, the inspection partial image, and the difference partial image. Whether the defect candidate at the selected defect candidate position D5 is the image defect D6 by inputting this feature amount into a machine learning model (multiple regression analysis, partial least squares method, neural network, support vector machine, ridge regression, etc.) Determine (classification) whether or not non-defective.
　第２検出処理の前処理として、基準画像Ｄ１及び検査画像Ｄ２の位置合わせ処理、ノイズ低減のための空間フィルタ処理、画像欠陥を強調するための空間フィルタ処理を実施してもよい。 As preprocessing of the second detection processing, alignment processing of the reference image D1 and the inspection image D2, spatial filter processing for noise reduction, and spatial filter processing for enhancing an image defect may be performed.
　第２検出処理は、基準部分画像、検査部分画像、及び差分部分画像を深層学習で生成された機械学習モデルに入力し、機械学習モデルによって画像欠陥Ｄ６を検出する処理であってもよい。 The second detection process may be a process of inputting the reference partial image, the inspection partial image, and the difference partial image into a machine learning model generated by deep learning, and detecting the image defect D6 by the machine learning model.
　また、第２検出処理は、基準部分画像、検査部分画像、及び差分部分画像に対してフーリエ変換処理、ウェーブレット変換等を用いた周波数解析処理、又は多重解像度解析処理を実施して基底画像を生成し、生成された基底画像から求めた特徴量から、画像欠陥Ｄ６を検出する処理であってもよい。 In the second detection process, a base image is generated by performing a Fourier transform process, a frequency analysis process using wavelet transform, or a multiple resolution analysis process on the reference partial image, the inspection partial image, and the difference partial image. Alternatively, the image defect D6 may be detected from the feature amount obtained from the generated base image.
　さらに、第２検出処理は、基準部分画像、検査部分画像、及び差分部分画像に対してフーリエ変換処理、ウェーブレット変換等を用いた周波数解析処理、又は多重解像度解析処理を実施して基底画像を生成し、生成された基底画像又は基底画像から求めた特徴量を機械学習モデル又は深層学習モデルに入力し、機械学習モデル又は深層学習モデルによって画像欠陥Ｄ６を検出する処理であってもよい。 Furthermore, in the second detection process, a base image is generated by performing Fourier transform processing, frequency analysis processing using wavelet transform or the like, or multiple resolution analysis processing on the reference partial image, inspection partial image, and difference partial image. The process may be a process of inputting the generated base image or a feature value obtained from the base image into a machine learning model or a deep learning model, and detecting the image defect D6 by the machine learning model or the deep learning model.
　それぞれの処理において、人間の視覚特性に合わせた量に回帰し、回帰結果に対して閾値を超えた場合に画像欠陥Ｄ６と判定してもよい。 In each process, the image may be regressed to an amount matched to human visual characteristics, and the image defect D6 may be determined when the threshold value is exceeded for the regression result.
　機械学習モデル又は深層学習モデルを使用する場合、モデルが人間の視覚特性をよく表現できることが望ましい。このため、モデルを作成するための学習データには、人による官能評価実験によって得られた官能値もしくはそれを加工したデータを用いることが望ましい。 When using a machine learning model or a deep learning model, it is desirable that the model can well express human visual characteristics. For this reason, as learning data for creating a model, it is desirable to use sensory values obtained by human sensory evaluation experiments or data obtained by processing the sensory values.
　第２検出処理は、以上の処理に限定されず、欠陥候補位置の欠陥候補が画像欠陥Ｄ６であるか否かの分類ができればよい。 The second detection process is not limited to the above process, as long as the defect candidate at the defect candidate position can be classified as to whether or not it is the image defect D6.
　なお、第２検出処理部２６は、複数の演算を並列に処理する不図示の並列演算器を備えていてもよい。並列演算器を用いることで、第２検出処理を並列に実行して高速化することができる。これにより、上限値Ｄ４を大きな値とすることができる。 The second detection processing unit 26 may include a parallel computing unit (not shown) that processes a plurality of operations in parallel. By using the parallel computing unit, it is possible to execute the second detection process in parallel to speed up the process. As a result, the upper limit value D4 can be made a large value.
　また、並列演算器は、例えば単一の基板に実装された複数のＣＰＵ（Central Processing Unit）を備えて構成される。並列演算器は、１又は複数のＣＰＵが実装された複数の基板を備えて構成されていてもよい。これにより、基板の数を増減させることで、並列演算器を構成するＣＰＵの数を増減させることができる。したがって、第２検出処理の実行に必要なリソースを有するハードウェアの構築が容易となる。 Also, the parallel computing unit is configured to include, for example, a plurality of CPUs (Central Processing Units) mounted on a single substrate. The parallel computing unit may be configured to include a plurality of substrates on which one or more CPUs are mounted. Thus, by increasing or decreasing the number of substrates, it is possible to increase or decrease the number of CPUs constituting the parallel computing device. Therefore, construction of hardware having resources necessary to execute the second detection process is facilitated.
　さらに、並列演算器は、ＣＰＵに代えて、又はＣＰＵと共に、複数のＧＰＵ（Graphics Processing Unit）を備えて構成されていてもよい。第１検出処理部２２及び第２検出処理部２６において、同じ並列演算器を共有してもよい。 Furthermore, the parallel computing unit may be configured to include a plurality of GPUs (Graphics Processing Units) instead of or together with the CPU. The first detection processing unit 22 and the second detection processing unit 26 may share the same parallel computing unit.
　〔第１検出処理と第２検出処理の演算量〕
　前述の前処理をそれぞれ含めた第１検出処理と第２検出処理とを比較すると、第２検出処理は相対的に複雑な検査を行うため、第２検出処理の単位検査あたりの演算量は第１の検出処理の単位検査あたりの演算量よりも大きい。即ち、画像全体に対する第２検出処理の演算量は、画像全体に対する第１検出処理の演算量よりも大きい。これは、仮に画像のすべての画素位置を欠陥候補位置として第２検出処理を実施した場合に、第１の検出処理と比較して演算量が大きいということを意味する。 [Calculation amount of first detection processing and second detection processing]
Comparing the first detection process and the second detection process including the above-described preprocessing, the second detection process performs a relatively complicated inspection, so the amount of calculation per unit inspection of the second detection process is This is larger than the amount of calculation per unit inspection of the 1 detection process. That is, the operation amount of the second detection process for the entire image is larger than the operation amount of the first detection process for the entire image. This means that if the second detection process is performed assuming that all pixel positions of the image are defect candidate positions, the amount of calculation is large compared to the first detection process.
　また、画像の単一領域に対する第２検出処理の演算量は、同じ単一領域に対する第１検出処理の演算量よりも大きい、とも表現することができる。なお、第２検出処理部２６に並列演算器を使用している場合は、画像全体で比較する方が望ましい。 In addition, it can also be expressed that the operation amount of the second detection processing for a single area of the image is larger than the operation amount of the first detection processing for the same single area. When a parallel computing unit is used for the second detection processing unit 26, it is preferable to compare the entire image.
　このように、第２検出処理の単位検査あたりの演算量が相対的に大きい。このため、画像全体に対して第２検出処理を行うと、演算量が膨大となり、画像欠陥検出処理の演算時間が目標時間内に収まらない。また、第１検出処理によって検出された欠陥候補位置に対してのみ第２検出処理を行う場合であっても、欠陥候補位置の数が多すぎる場合には、同様に画像欠陥検出処理の演算時間が目標時間内に収まらなくなることがあり得る。そこで、第２検出処理部２６は、第１検出処理により検出された欠陥候補位置のうち、予め定められた上限値Ｄ４以下の数の欠陥候補位置についてのみ第２検出処理を行う。これにより、画像欠陥検出処理を目標時間内に終了させることができる。 Thus, the amount of operation per unit inspection of the second detection process is relatively large. For this reason, when the second detection process is performed on the entire image, the amount of calculation becomes enormous, and the calculation time of the image defect detection process does not fall within the target time. In addition, even when the second detection process is performed only on the defect candidate position detected by the first detection process, if the number of defect candidate positions is too large, the calculation time of the image defect detection process similarly May not fit within the target time. Therefore, the second detection processing unit 26 performs the second detection process only on defect candidate positions having a number equal to or less than a predetermined upper limit value D4 among the defect candidate positions detected by the first detection process. Thereby, the image defect detection process can be completed within the target time.
　〔上限値〕
　１枚の印刷物の画像欠陥を検出するための目標時間をＴＧ、１枚の印刷物の第１検出処理にかかる時間をＴ１、１箇所の欠陥候補位置について第２検出処理に係る時間をＴ２とすると、上限値Ｄ４は以下の式１を満たす必要がある。 〔upper limit〕
Target time for detecting an image defect of one printed matter T G , time taken for the first detection process of one printed matter T 1 , time taken for the second detection process for one defect candidate position T Assuming that it is 2 , the upper limit value D4 needs to satisfy the following equation 1.
　Ｄ４＜（ＴＧ－Ｔ１）／Ｔ２　…（式１）
　選択部２４において使用する上限値Ｄ４は、式１を満たす整数のうち最大の整数とすればよい。 D4 <(T G -T 1 ) / T 2 (Equation 1)
The upper limit value D4 used in the selection unit 24 may be the largest integer among the integers satisfying Expression 1.
　なお、時間Ｔ１は第１検出処理部２２の処理能力によって決まる時間である。また、時間Ｔ２は第２検出処理部２６の処理能力によって決まる時間である。また、時間Ｔ１及び時間Ｔ２は、画像欠陥検出対象の画像のサイズ及び複雑度によって変化する場合がある。なお、本実施形態では、時間Ｔ１は画像に依存せずに決まる値である。 The time T 1 is a time determined by the processing capability of the first detection processing section 22. The time T 2 is a time determined by the processing capability of the second detection processing section 26. The time T 1 and time T 2 are, may vary by the size and complexity of the image defect detection target image. In the present embodiment, the time T 1 is a value determined without depending on the image.
　以上のように、第１の実施形態に係る画像欠陥検出装置１０によれば、第２検出処理を行う欠陥候補位置の数の上限値を、検出時間の目標時間、第１検出処理部２２の処理能力、及び第２検出処理部２６の処理能力に応じて合わせて設定することができるので、高精度な第２検出処理の処理時間内での実施を保証することができる。また、演算時間よりも性能を優先して第２検出処理を設計することができるため、第２検出処理を複雑化及び高性能化することが可能となる。 As described above, according to the image defect detection apparatus 10 according to the first embodiment, the upper limit value of the number of defect candidate positions on which the second detection process is performed, the target time of the detection time, Since it can be set according to the processing capacity and the processing capacity of the second detection processing unit 26, it is possible to guarantee the implementation of the highly accurate second detection processing within the processing time. In addition, since the second detection process can be designed with priority given to performance over operation time, it is possible to make the second detection process complicated and have high performance.
　したがって、画像欠陥検出処理を適切に行いつつ、目標時間内に終了させることができる。 Therefore, it is possible to complete the process within the target time while properly performing the image defect detection process.
　＜第２の実施形態＞
　〔画像欠陥検出装置の構成〕
　図３は、第２の実施形態に係る画像欠陥検出装置１２の内部構成を示すブロック図である。なお、図１に示すブロック図と共通する部分には同一の符号を付し、その詳細な説明は省略する。 Second Embodiment
[Configuration of Image Defect Detection Device]
FIG. 3 is a block diagram showing an internal configuration of the image defect detection device 12 according to the second embodiment. In addition, the same code | symbol is attached | subjected to the part which is common in the block diagram shown in FIG. 1, and the detailed description is abbreviate | omitted.
　画像欠陥検出装置１２は、印刷物の画像欠陥を検出する装置である。図３に示すように、画像欠陥検出装置１２は、判断部２８及び良否判定部３０を備えている点が画像欠陥検出装置１０と異なる。 The image defect detection device 12 is a device that detects an image defect of a printed matter. As shown in FIG. 3, the image defect detection device 12 differs from the image defect detection device 10 in that the image defect detection device 12 includes a determination unit 28 and a quality determination unit 30.
　判断部２８は、第１検出処理により検出された欠陥候補位置のうち、選択部２４によって選択されなかった欠陥候補位置に関する判断を行う。 The determination unit 28 determines a defect candidate position not selected by the selection unit 24 among the defect candidate positions detected by the first detection process.
　良否判定部３０は、第２検出処理部２６の検出結果及び判断部２８の判断結果に基づいて、印刷物の良否判定を行う。 The quality determination unit 30 determines the quality of the printed matter based on the detection result of the second detection processing unit 26 and the determination result of the determination unit 28.
　〔画像欠陥検出方法〕
　画像欠陥検出装置１２における画像欠陥検出方法について説明する。図４は、画像欠陥検出方法の処理と各処理に用いるデータについて示したフローチャートである。なお、図２に示すフローチャートと共通する部分には同一の符号を付し、その詳細な説明は省略する。 [Image defect detection method]
The image defect detection method in the image defect detection apparatus 12 will be described. FIG. 4 is a flowchart showing processing of the image defect detection method and data used for each processing. In addition, the same code | symbol is attached | subjected to the part which is common in the flowchart shown in FIG. 2, and the detailed description is abbreviate | omitted.
　ステップＳ１において、画像取得部２０により基準画像Ｄ１及び検査画像Ｄ２を取得する。 In step S1, the reference image D1 and the inspection image D2 are acquired by the image acquisition unit 20.
　続いて、ステップＳ２では、第１検出処理部２２において第１検出処理を行い、検査画像Ｄ２の欠陥候補位置Ｄ３を検出する。 Subsequently, in step S2, the first detection processing unit 22 performs a first detection process to detect a defect candidate position D3 of the inspection image D2.
　また、第１検出処理部２２は、欠陥候補位置Ｄ３の強度値を検出する。欠陥候補位置Ｄ３の強度値とは、画像欠陥の候補の視認されやすさを示す指標である。欠陥候補位置Ｄ３の強度値は、強度が大きい、即ち画像欠陥の候補が視認されやすいほど相対的に大きい値となり、強度が小さい、即ち視認されにくいほど相対的に小さい値となる。 In addition, the first detection processing unit 22 detects the intensity value of the defect candidate position D3. The intensity value of the defect candidate position D3 is an index indicating the visibility of the image defect candidate. The intensity value of the defect candidate position D3 is relatively large as the intensity is high, that is, as the image defect candidate is easy to be recognized, and as the intensity is low, ie as it is difficult to be visually recognized.
　次に、ステップＳ１１において、選択部２４は、複数の欠陥候補位置Ｄ３の中から上限値Ｄ４以下の数の選択欠陥候補位置Ｄ５を選択する。ここでは、一例として、欠陥候補位置Ｄ３の数が２０箇所であり、上限値Ｄ４が１５箇所であるとする。 Next, in step S11, the selection unit 24 selects the number of selected defect candidate positions D5 equal to or less than the upper limit value D4 from among the plurality of defect candidate positions D3. Here, as an example, it is assumed that the number of defect candidate positions D3 is 20 and the upper limit value D4 is 15.
　本実施形態では、選択部２４は、強度が弱い欠陥候補位置Ｄ３を優先して選択する。即ち、複数の欠陥候補位置Ｄ３を強度が弱い順にソートし、強度の弱い順に１番目から１５番目までの１５箇所を選択欠陥候補位置Ｄ５として選択する。選択された選択欠陥候補位置Ｄ５は、第２検出処理部２６に入力される。 In the present embodiment, the selection unit 24 preferentially selects the defect candidate position D3 whose intensity is weak. That is, the plurality of defect candidate positions D3 are sorted in the order of weak intensity, and the 15 first to fifteenth positions are selected as the selected defect candidate position D5 in ascending order of intensity. The selected selected defect candidate position D5 is input to the second detection processing unit 26.
　また、選択部２４は、複数の欠陥候補位置Ｄ３のうち選択欠陥候補位置Ｄ５以外の欠陥候補位置を非選択欠陥候補位置Ｄ１１とする。即ち、強度の弱い順に１６番目から２０番目までの５箇所が非選択欠陥候補位置Ｄ１１となる。非選択欠陥候補位置Ｄ１１は、判断部２８に入力される。 Further, the selection unit 24 sets a defect candidate position other than the selected defect candidate position D5 among the plurality of defect candidate positions D3 as a non-selected defect candidate position D11. That is, five non-selected defect candidate positions D11 from the sixteenth to the twentieth places in the order of weak intensity. The non-selected defect candidate position D11 is input to the determination unit 28.
　続くステップＳ１２では、判断部２８は、入力された非選択欠陥候補位置Ｄ１１について、画像欠陥が存在すると判断する。ここでは、５箇所の非選択欠陥候補位置Ｄ１１に画像欠陥Ｄ６が存在するという判断結果Ｄ１２を出力する。 In the subsequent step S12, the determination unit 28 determines that an image defect exists at the input non-selected defect candidate position D11. Here, the determination result D12 that the image defect D6 exists at five non-selected defect candidate positions D11 is output.
　次に、ステップＳ４では、第２検出処理部２６は、１５箇所の選択欠陥候補位置Ｄ５に対して画像欠陥Ｄ６を検出する第２検出処理を行う。 Next, in step S4, the second detection processing unit 26 performs a second detection process of detecting the image defect D6 with respect to 15 selected defect candidate positions D5.
　最後に、ステップＳ１３において、良否判定部３０は、判断部２８の判断結果Ｄ１２による画像欠陥Ｄ６と第２検出処理部２６の検出結果である画像欠陥Ｄ６に基づいて、印刷物の良否判定を行い、良否判定結果Ｄ１３を出力する。例えば、画像欠陥Ｄ６の数の良否判定閾値を１０箇所として良否判定する。この場合は、判断部２８の判断結果Ｄ１２による画像欠陥Ｄ６の数と第２検出処理部２６の検出結果である画像欠陥Ｄ６の和が１０箇所未満であれば良品印刷物（合格印刷物）、１０箇所以上であれば不良品印刷物（不合格印刷物）と判定する。 Finally, in step S13, the good or defective judgment unit 30 judges the quality of the printed matter based on the image defect D6 by the judgment result D12 of the judgment unit 28 and the image defect D6 which is the detection result of the second detection processing unit 26. The pass / fail judgment result D13 is output. For example, the pass / fail judgment is performed with ten pass / fail judgment thresholds of the number of image defects D6. In this case, if the sum of the number of image defects D6 according to the determination result D12 of the determination unit 28 and the image defect D6 which is the detection result of the second detection processing unit 26 is less than 10, non-defective printed matter (accepted printed matter) If it is above, it will be judged as inferior goods printed matter (rejected printed matter).
　以上で、画像欠陥検出が終了する。 This is the end of the image defect detection.
　第２の実施形態に係る画像欠陥検出装置１２によれば、強度が弱い欠陥候補位置を優先して第２検出処理を行い、第２検出処理を行わない欠陥候補位置について画像欠陥が存在すると判断することで、高精度な欠陥検出として、人の視覚系に対応させるような微妙な判断を行うことが容易となる。したがって、画像欠陥検出処理を適切に行いつつ、目標時間内に終了させることができる。 According to the image defect detection apparatus 12 according to the second embodiment, the second detection process is performed by giving priority to defect candidate positions having weak intensities, and it is determined that an image defect exists at a defect candidate position where the second detection process is not performed. By doing so, it becomes easy to make subtle judgments that make it correspond to the human visual system as highly accurate defect detection. Therefore, it is possible to complete the process within the target time while properly performing the image defect detection process.
　なお、ステップＳ１１において選択欠陥候補位置Ｄ５を選択した結果、非選択欠陥候補位置Ｄ１１の数が良否判定閾値を超えた場合は、直ちに不良品印刷物と判定して第２検出処理を省略してもよい。 As a result of selecting the selected defect candidate position D5 in step S11, if the number of non-selected defect candidate positions D11 exceeds the pass / fail judgment threshold, it is immediately judged as a defective printed matter and the second detection process is omitted. Good.
　＜第３の実施形態＞
　〔画像欠陥検出方法〕
　図５は、画像欠陥検出装置１２における第３の実施形態に係る画像欠陥検出方法の処理と各処理に用いるデータについて示したフローチャートである。なお、図４に示すフローチャートと共通する部分には同一の符号を付し、その詳細な説明は省略する。 Third Embodiment
[Image defect detection method]
FIG. 5 is a flowchart showing the process of the image defect detection method according to the third embodiment in the image defect detection apparatus 12 and the data used for each process. The parts common to the flowchart shown in FIG. 4 are denoted by the same reference numerals, and the detailed description thereof is omitted.
　まず、ステップＳ１において、画像取得部２０により基準画像Ｄ１及び検査画像Ｄ２を取得する。また、ステップＳ２において、検査画像Ｄ２の欠陥候補位置Ｄ３及びその強度（強度値）を検出する。 First, in step S1, the reference image D1 and the inspection image D2 are acquired by the image acquisition unit 20. Further, in step S2, the defect candidate position D3 of the inspection image D2 and its intensity (intensity value) are detected.
　次に、ステップＳ２１において、選択部２４は、複数の欠陥候補位置Ｄ３の中から上限値Ｄ４以下の数の選択欠陥候補位置Ｄ５を選択する。ここでは、第２の実施形態と同様に、欠陥候補位置Ｄ３の数が２０箇所であり、上限値Ｄ４が１５箇所であるとする。選択部は、１５箇所の選択欠陥候補位置Ｄ５を選択する。 Next, in step S21, the selection unit 24 selects the number of selected defect candidate positions D5 equal to or less than the upper limit value D4 from among the plurality of defect candidate positions D3. Here, as in the second embodiment, it is assumed that the number of defect candidate positions D3 is 20 and the upper limit D4 is 15. The selection unit selects fifteen selected defect candidate positions D5.
　本実施形態では、選択部２４は、強度が強い欠陥候補位置Ｄ３を優先して選択する。即ち、複数の欠陥候補位置Ｄ３を強度が強い順にソートし、強度の強い順に１番目から１５番目までの１５箇所を選択欠陥候補位置Ｄ５として選択する。選択された選択欠陥候補位置Ｄ５は、第２検出処理部２６に入力される。 In the present embodiment, the selection unit 24 preferentially selects the defect candidate position D3 having a high strength. That is, the plurality of defect candidate positions D3 are sorted in descending order of strength, and the 15 first to fifteenth positions are selected as the selected defect candidate positions D5 in descending order of strength. The selected selected defect candidate position D5 is input to the second detection processing unit 26.
　また、選択部２４は、複数の欠陥候補位置Ｄ３のうち選択欠陥候補位置Ｄ５以外の欠陥候補位置を非選択欠陥候補位置Ｄ１１とする。即ち、強度の強い順に１６番目から２０番目までの５箇所が非選択欠陥候補位置Ｄ１１となる。非選択欠陥候補位置Ｄ１１は、判断部２８に入力される。 Further, the selection unit 24 sets a defect candidate position other than the selected defect candidate position D5 among the plurality of defect candidate positions D3 as a non-selected defect candidate position D11. That is, five non-selected defect candidate positions D11 are the sixteenth to twentieth places in descending order of strength. The non-selected defect candidate position D11 is input to the determination unit 28.
　続くステップＳ２２では、判断部２８は、入力された非選択欠陥候補位置Ｄ１１について、画像欠陥が存在しないと判断する。ここでは、５箇所の非選択欠陥候補位置Ｄ１１に画像欠陥が存在しないという判断結果Ｄ１２を出力する。 In the subsequent step S22, the determination unit 28 determines that there is no image defect at the input non-selected defect candidate position D11. Here, a determination result D12 indicating that no image defect exists at five non-selected defect candidate positions D11 is output.
　次に、ステップＳ４では、第２検出処理部２６は、１５箇所の選択欠陥候補位置Ｄ５に対して画像欠陥Ｄ６を検出する。 Next, in step S4, the second detection processing unit 26 detects the image defect D6 at 15 selected defect candidate positions D5.
　最後に、ステップＳ１２において、良否判定部３０は、第２検出処理部２６の検出結果である画像欠陥Ｄ６に基づいて、印刷物の良否判定を行い、良否判定結果Ｄ１３を出力する。 Finally, in step S12, the good or bad judgment unit 30 judges the quality of the printed matter based on the image defect D6 which is the detection result of the second detection processing unit 26, and outputs the good / bad judgment result D13.
　以上で、画像欠陥検出が終了する。 This is the end of the image defect detection.
　第３の実施形態に係る画像欠陥検出装置１２によれば、強度が強い欠陥候補位置を優先して第２検出処理を行い、第２検出処理を行わない欠陥候補位置について画像欠陥が存在しないと判断することで、画像欠陥検出処理を適切に行いつつ、目標時間内に終了させることができる。本実施形態では、１箇所でも画像欠陥が存在する場合には不良品印刷物と判定する場合に有効である。 According to the image defect detection apparatus 12 according to the third embodiment, the second detection process is performed with priority given to the defect candidate position having high intensity, and the image defect does not exist at the defect candidate position where the second detection process is not performed. By making the determination, it is possible to complete the process within the target time while appropriately performing the image defect detection process. In the present embodiment, it is effective in the case where it is determined as a defective product printed matter if an image defect exists even in one place.
　本実施形態では、非選択欠陥候補位置Ｄ１１について、画像欠陥が存在しないと判断したが、判断部２８による判断を行わなくてもよい。この場合、画像欠陥検出装置１２は、判断部２８を備えなくてもよい。 In the present embodiment, it is determined that the image defect does not exist at the non-selected defect candidate position D11, but the determination by the determination unit 28 may not be performed. In this case, the image defect detection device 12 may not include the determination unit 28.
　なお、ステップＳ２１において、選択部２４の選択した選択欠陥候補位置Ｄ５の数が良否判定閾値に満たない場合は、直ちに良品印刷物と判定して第２検出処理を省略してもよい。 If the number of selected defect candidate positions D5 selected by the selection unit 24 does not satisfy the pass / fail judgment threshold in step S21, the second detection process may be omitted by immediately judging as a non-defective printed matter.
　＜第４の実施形態＞
　図６は、第４の実施形態に係る画像欠陥検出装置１４の内部構成を示すブロック図である。なお、図１に示すブロック図と共通する部分には同一の符号を付し、その詳細な説明は省略する。 Fourth Embodiment
FIG. 6 is a block diagram showing an internal configuration of the image defect detection device 14 according to the fourth embodiment. In addition, the same code | symbol is attached | subjected to the part which is common in the block diagram shown in FIG. 1, and the detailed description is abbreviate | omitted.
　図６に示すように、画像欠陥検出装置１４の選択部２４は、格納部３２を備えている。格納部３２は、第１検出処理部２２が検出した欠陥候補位置を待ち行列の末尾に格納する。また、選択部２４は、欠陥候補位置を待ち行列の先頭から選択する。即ち、選択部２４は、欠陥候補位置を格納部３２に格納された正順に選択する、いわゆるキューとして機能する。 As shown in FIG. 6, the selection unit 24 of the image defect detection device 14 includes a storage unit 32. The storage unit 32 stores the defect candidate position detected by the first detection processing unit 22 at the end of the queue. Further, the selection unit 24 selects a defect candidate position from the head of the queue. That is, the selection unit 24 functions as a so-called queue that selects defect candidate positions in the order of being stored in the storage unit 32.
　選択部２４は、欠陥候補位置Ｄ３を格納部３２に格納された順に選択欠陥候補位置Ｄ５として選択する。選択した選択欠陥候補位置Ｄ５の数が上限値Ｄ４に達すると、残りの選択されていない欠陥候補位置Ｄ３を非選択欠陥候補位置Ｄ１１とする。 The selection unit 24 selects the defect candidate position D3 as the selected defect candidate position D5 in the order stored in the storage unit 32. When the number of selected selected defect candidate positions D5 reaches the upper limit D4, the remaining non-selected defect candidate positions D3 are set as non-selected defect candidate positions D11.
　第２の実施形態では、選択部２４は、強度が弱い欠陥候補位置Ｄ３を優先して選択した。また、第３の実施形態では、選択部２４は、強度が強い欠陥候補位置Ｄ３を優先して選択した。 In the second embodiment, the selection unit 24 preferentially selects the defect candidate position D3 whose intensity is weak. Further, in the third embodiment, the selection unit 24 selects the defect candidate position D3 with high strength in priority.
　しかしながら、これらの優先順位付けにはソート処理等の演算リソースを別途使用する。このため、優先順位付けを行わない場合と比較して選択する欠陥候補位置Ｄ３の上限値Ｄ４を小さくする必要性が生じる。したがって、なるべく多くの欠陥候補位置Ｄ３に対して第２検出処理を行うという観点を重要視する場合には、優先順位付けを行わない形態が好ましい。 However, calculation resources such as sorting are separately used for prioritizing these. For this reason, it is necessary to reduce the upper limit value D4 of the defect candidate position D3 to be selected as compared with the case where the prioritization is not performed. Therefore, when importance is attached to the viewpoint of performing the second detection process on as many defect candidate positions D3 as possible, it is preferable that priority is not assigned.
　本実施形態によれば、欠陥候補位置Ｄ３が格納された順に選択欠陥候補位置Ｄ５として選択するため、優先順位付けを行うことがなく、上限値Ｄ４を大きくすることができる。 According to the present embodiment, since the selected defect candidate positions D5 are selected in the order in which the defect candidate positions D3 are stored, the upper limit value D4 can be increased without performing prioritization.
　なお、選択部２４は、欠陥候補位置を待ち行列の末尾から選択してもよい。即ち、選択部２４は、欠陥候補位置を格納部３２に格納された逆順に選択する、いわゆるスタックとして機能する。このように構成しても、優先順位付けを行うことがないため、上限値Ｄ４を大きくすることができる。 The selection unit 24 may select the defect candidate position from the end of the queue. That is, the selection unit 24 functions as a so-called stack that selects defect candidate positions in the reverse order stored in the storage unit 32. Even with this configuration, since the prioritization is not performed, the upper limit value D4 can be increased.
　＜第５の実施形態＞
　〔画像欠陥検出装置の構成〕
　図７は、第５の実施形態に係る画像欠陥検出装置１６の内部構成を示すブロック図である。なお、図３に示すブロック図と共通する部分には同一の符号を付し、その詳細な説明は省略する。 Fifth Embodiment
[Configuration of Image Defect Detection Device]
FIG. 7 is a block diagram showing an internal configuration of the image defect detection apparatus 16 according to the fifth embodiment. The same reference numerals as in the block diagram shown in FIG. 3 denote the same parts in FIG.
　図７に示すように、画像欠陥検出装置１６は、通知部３４及び設定部３６を備えている。 As shown in FIG. 7, the image defect detection device 16 includes a notification unit 34 and a setting unit 36.
　通知部３４は、欠陥候補位置の数が上限値を超えた場合にユーザに警告又はエラーを通知する。通知部３４は、不図示の表示部に上限値を超えた旨の表示を行ってもよいし、不図示のスピーカからメッセージを発してもよいし、不図示のライトを点灯してもよい。 The notification unit 34 notifies the user of a warning or an error when the number of defect candidate positions exceeds the upper limit value. The notification unit 34 may display an indication that the upper limit value is exceeded on the display unit (not shown), may issue a message from a speaker (not shown), or may turn on a light (not shown).
　設定部３６は、通知部３４による通知の有無を設定する。ユーザは、不図示の入力インターフェースを用いて、通知の有無を入力することができる。入力された情報は、設定部３６に設定される。 The setting unit 36 sets the presence or absence of notification by the notification unit 34. The user can input the presence or absence of notification using an input interface (not shown). The input information is set in the setting unit 36.
　通知部３４は、設定部３６において通知有りに設定され、かつ欠陥候補位置の数が上限値を超えた場合にユーザに通知する。 The notification unit 34 notifies the user when the notification is set in the setting unit 36 and the number of defect candidate positions exceeds the upper limit value.
　〔画像欠陥検出方法〕
　画像欠陥検出装置１６における画像欠陥検出方法について説明する。図８は、画像欠陥検出方法の処理と各処理に用いるデータについて示したフローチャートである。なお、図４に示すフローチャートと共通する部分には同一の符号を付し、その詳細な説明は省略する。 [Image defect detection method]
The image defect detection method in the image defect detection apparatus 16 will be described. FIG. 8 is a flowchart showing processing of the image defect detection method and data used for each processing. The parts common to the flowchart shown in FIG. 4 are denoted by the same reference numerals, and the detailed description thereof is omitted.
　最初に、ステップＳ３１において、ユーザは、不図示の入力インターフェースを用いて、通知部３４による通知の有無を入力する。設定部３６は、入力された情報に基づいて通知部３４による通知の有無を設定する。 First, in step S31, the user inputs the presence or absence of notification by the notification unit 34 using an input interface (not shown). The setting unit 36 sets the presence or absence of notification by the notification unit 34 based on the input information.
　次に、ステップＳ１において、画像取得部２０により基準画像Ｄ１及び検査画像Ｄ２を取得する。また、ステップＳ２において、第１検出処理部２２により第１検出処理を行い、検査画像Ｄ２の欠陥候補位置Ｄ３及びその強度を検出する。欠陥候補位置Ｄ３は通知部３４に入力される。 Next, in step S1, the reference image D1 and the inspection image D2 are acquired by the image acquisition unit 20. In step S2, the first detection processing unit 22 performs the first detection process to detect the defect candidate position D3 of the inspection image D2 and the intensity thereof. The defect candidate position D3 is input to the notification unit 34.
　続くステップＳ３２では、通知部３４は、不図示のメモリに記憶された上限値Ｄ４を取得する。さらに、通知部３４は、この上限値Ｄ４と第１検出処理部２２から入力された欠陥候補位置Ｄ３の数とを比較し、欠陥候補位置Ｄ３の数が上限値Ｄ４を超えているか否かを判定する。 In the subsequent step S32, the notification unit 34 acquires an upper limit value D4 stored in a memory (not shown). Furthermore, the notification unit 34 compares the upper limit value D4 with the number of defect candidate positions D3 input from the first detection processing unit 22, and determines whether the number of defect candidate positions D3 exceeds the upper limit value D4. judge.
　欠陥候補位置Ｄ３の数が上限値Ｄ４を超えている場合は、ステップＳ３３に移行する。一方、欠陥候補位置Ｄ３の数が上限値Ｄ４を超えていない場合は、通知を行わずにステップＳ１１に移行する。 If the number of defect candidate positions D3 exceeds the upper limit value D4, the process proceeds to step S33. On the other hand, when the number of defect candidate positions D3 does not exceed the upper limit value D4, the process proceeds to step S11 without performing the notification.
　ステップＳ３３では、通知部３４は、設定部３６によって通知有りに設定されているか否かを判定する。通知有りに設定されている場合は、ステップＳ３４に移行する。また、通知無しに設定されている場合は、通知を行わずにステップＳ１１に移行する。 In step S33, the notification unit 34 determines whether the setting unit 36 has set notification. If the notification is set, the process proceeds to step S34. Moreover, when it sets to no notification, it transfers to step S11, without performing notification.
　ステップＳ３４では、通知部３４は、欠陥候補位置Ｄ３の数が上限値Ｄ４を超えた旨の警告又はエラーをユーザに通知する。続いて、ステップＳ１１に移行する。 In step S34, the notification unit 34 notifies the user of a warning or an error that the number of defect candidate positions D3 has exceeded the upper limit value D4. Then, it transfers to step S11.
　ステップＳ１１では、選択部２４は、複数の欠陥候補位置Ｄ３の中から上限値Ｄ４以下の数の選択欠陥候補位置Ｄ５を選択する。これ以降の処理は、第２の実施形態と同様である。 In step S11, the selection unit 24 selects the number of selected defect candidate positions D5 equal to or less than the upper limit value D4 from among the plurality of defect candidate positions D3. The subsequent processing is the same as that of the second embodiment.
　このように、欠陥候補位置Ｄ３の数が上限値Ｄ４を超えた場合にユーザに通知することで、ユーザがその旨を知ることができる。 As described above, when the number of defect candidate positions D3 exceeds the upper limit value D4, the user can know that by notifying the user.
　＜第６の実施形態＞
　画像欠陥検出装置１０、１２、１４、及び１６は、印刷装置に適用することが可能である。ここでは、画像欠陥検出装置１２を適用したインクジェット印刷装置について説明する。 Sixth Embodiment
The image defect detection devices 10, 12, 14 and 16 can be applied to a printing device. Here, an inkjet printing apparatus to which the image defect detection device 12 is applied will be described.
　〔インクジェット印刷装置の構成〕
　図９は、インクジェット印刷装置１００の全体の概略構成を示す全体構成図である。図９に示すように、インクジェット印刷装置１００は、印刷媒体である枚葉の用紙Ｐにシアン（Ｃ）、マゼンタ（Ｍ）、イエロー（Ｙ）、及びクロ（Ｋ）の４色のインクを吐出してカラー画像を印刷する印刷機である。 [Configuration of inkjet printing apparatus]
FIG. 9 is an overall configuration diagram showing an overall schematic configuration of the inkjet printing apparatus 100. As shown in FIG. As shown in FIG. 9, the inkjet printing apparatus 100 ejects four color inks of cyan (C), magenta (M), yellow (Y), and black (K) onto a sheet of paper P as a printing medium. Printing machine that prints color images.
　用紙Ｐには汎用の印刷用紙が使用される。汎用の印刷用紙とは、いわゆるインクジェット専用紙ではなく、一般のオフセット印刷などに用いられる塗工紙などのセルロースを主体とした用紙をいう。また、インクには水性インクが使用される。水性インクとは、水及び水に可溶な溶媒に染料、顔料などの色材を溶解又は分散させたインクをいう。 As the paper P, general-purpose printing paper is used. A general-purpose printing sheet is not a so-called inkjet-only sheet but a sheet mainly composed of cellulose such as a coated sheet used for general offset printing. In addition, an aqueous ink is used as the ink. The aqueous ink refers to an ink in which a coloring material such as a dye or a pigment is dissolved or dispersed in water and a solvent soluble in water.
　図９に示すように、インクジェット印刷装置１００は、搬送部１１０、印刷部１２０、撮像部１３０、乾燥部１４０、選別部１５０、及び排紙部１６０等を備えて構成される。 As shown in FIG. 9, the inkjet printing apparatus 100 includes a conveyance unit 110, a printing unit 120, an imaging unit 130, a drying unit 140, a sorting unit 150, a paper discharge unit 160, and the like.
　〔搬送部〕
　搬送部１１０は、不図示の給紙部から給紙された用紙Ｐを搬送方向（Ｙ方向）に搬送する。搬送部１１０は、上流側プーリ１１２、下流側プーリ１１４、及び搬送ベルト１１６を備えている。 [Transporter]
The transport unit 110 transports the sheet P fed from a paper feed unit (not shown) in the transport direction (Y direction). The conveyance unit 110 includes an upstream pulley 112, a downstream pulley 114, and a conveyance belt 116.
　上流側プーリ１１２は、水平方向に延びる不図示の回転軸を有し、回転軸が回転自在に軸支されている。下流側プーリ１１４は、上流側プーリ１１２の回転軸と平行な不図示の回転軸を有し、回転軸が回転自在に軸支されている。 The upstream pulley 112 has a rotating shaft (not shown) extending in the horizontal direction, and the rotating shaft is rotatably supported. The downstream pulley 114 has a rotating shaft (not shown) parallel to the rotating shaft of the upstream pulley 112, and the rotating shaft is rotatably supported.
　搬送ベルト１１６は、ステンレス製の無端状のベルトである。搬送ベルト１１６は、上流側プーリ１１２及び下流側プーリ１１４に架け渡されている。ステンレス製の搬送ベルト１１６を使用することで、用紙Ｐの平坦性を良好に保つことができる。 The conveying belt 116 is an endless belt made of stainless steel. The transport belt 116 is stretched over the upstream pulley 112 and the downstream pulley 114. By using the conveying belt 116 made of stainless steel, the flatness of the sheet P can be maintained well.
　下流側プーリ１１４は、駆動手段として不図示のモータを有している。モータが駆動すると、下流側プーリ１１４が図９において左回りに回転する。上流側プーリ１１２は、下流側プーリ１１４の回転に従動して図９において左回りに回転する。上流側プーリ１１２及び下流側プーリ１１４の回転により、搬送ベルト１１６は上流側プーリ１１２及び下流側プーリ１１４の間を走行経路に沿って走行する。 The downstream pulley 114 has a motor (not shown) as drive means. When the motor is driven, the downstream pulley 114 rotates counterclockwise in FIG. The upstream pulley 112 rotates counterclockwise in FIG. 9 following the rotation of the downstream pulley 114. The rotation of the upstream pulley 112 and the downstream pulley 114 causes the transport belt 116 to travel along the traveling path between the upstream pulley 112 and the downstream pulley 114.
　搬送ベルト１１６の搬送面には、不図示の給紙部から供給された用紙Ｐが載置される。搬送部１１０は、搬送ベルト１１６に載置された用紙Ｐを上流側プーリ１１２から下流側プーリ１１４に向かう搬送経路に沿って搬送し、排紙部１６０に受け渡す。この搬送経路の、印刷部１２０、撮像部１３０、乾燥部１４０、及び選別部１５０に対向する位置において、用紙Ｐは印刷面を水平に保持されて搬送される。 The sheet P supplied from a sheet feeding unit (not shown) is placed on the conveyance surface of the conveyance belt 116. The conveyance unit 110 conveys the sheet P placed on the conveyance belt 116 along a conveyance path from the upstream pulley 112 to the downstream pulley 114 and delivers the sheet P to the paper discharge unit 160. The sheet P is conveyed with the printing surface held horizontally at a position facing the printing unit 120, the imaging unit 130, the drying unit 140, and the sorting unit 150 in the transport path.
　搬送ベルト１１６に不図示の複数の吸着孔を設け、不図示のポンプにより搬送ベルト１１６の吸着孔を吸引することで、搬送ベルト１１６の搬送面に載置された用紙Ｐを搬送面に吸着保持してもよい。 A plurality of suction holes (not shown) are provided in the conveyance belt 116, and the suction holes of the conveyance belt 116 are suctioned by a pump (not shown), whereby the sheet P placed on the conveyance surface of the conveyance belt 116 is adsorbed and held on the conveyance surface. You may
　〔印刷部〕
　印刷部１２０は、用紙Ｐに画像を形成（印刷）する。印刷部１２０は、インクジェットヘッド１２２Ｃ、１２２Ｍ、１２２Ｙ、及び１２２Ｋを備えている。インクジェットヘッド１２２Ｃは、シアン（Ｃ）のインク滴をインクジェット方式で吐出する。同様に、インクジェットヘッド１２２Ｍ、１２２Ｙ、及び１２２Ｋは、それぞれマゼンタ（Ｍ）、イエロー（Ｙ）、及びクロ（Ｋ）のインク滴をインクジェット方式で吐出する。 [Printing department]
The printing unit 120 forms (prints) an image on the sheet P. The printing unit 120 includes inkjet heads 122C, 122M, 122Y, and 122K. The ink jet head 122C ejects ink droplets of cyan (C) by the ink jet method. Similarly, the inkjet heads 122M, 122Y, and 122K eject ink droplets of magenta (M), yellow (Y), and black (K) by an inkjet method, respectively.
　インクジェットヘッド１２２Ｃ、１２２Ｍ、１２２Ｙ及び１２２Ｋは、搬送ベルト１１６による用紙Ｐの搬送経路に沿って一定の間隔をもって配置される。インクジェットヘッド１２２Ｃ、１２２Ｍ、１２２Ｙ及び１２２Ｋは、それぞれラインヘッドで構成され、最大の用紙幅に対応する長さで形成される。インクジェットヘッド１２２Ｃ、１２２Ｍ、１２２Ｙ及び１２２Ｋは、ノズル面（ノズルが配列される面）が搬送ベルト１１６に対向するように配置される。 The inkjet heads 122 C, 122 M, 122 Y and 122 K are disposed at regular intervals along the transport path of the sheet P by the transport belt 116. The inkjet heads 122C, 122M, 122Y, and 122K are respectively line heads, and are formed to have a length corresponding to the maximum sheet width. The inkjet heads 122C, 122M, 122Y and 122K are arranged such that the nozzle surfaces (surfaces on which the nozzles are arranged) face the transport belt 116.
　インクジェットヘッド１２２Ｃ、１２２Ｍ、１２２Ｙ及び１２２Ｋは、ノズル面に形成されたノズルから、搬送ベルト１１６によって搬送される用紙Ｐに向けてインク滴を吐出することにより、用紙Ｐの印刷面に画像を形成する。 The inkjet heads 122C, 122M, 122Y, and 122K form an image on the printing surface of the sheet P by discharging ink droplets from the nozzles formed on the nozzle surface toward the sheet P conveyed by the conveyance belt 116. .
　〔撮像部〕
　撮像部１３０は、用紙Ｐの印刷面の画像を取得する。撮像部１３０は、用紙Ｐの搬送方向に対して印刷部１２０の下流側に配置される。撮像部１３０は、スキャナ１３２を備えている。 [Imaging unit]
The imaging unit 130 acquires an image of the printing surface of the sheet P. The imaging unit 130 is disposed downstream of the printing unit 120 with respect to the conveyance direction of the sheet P. The imaging unit 130 includes a scanner 132.
　スキャナ１３２は、インクジェットヘッド１２２Ｃ、１２２Ｍ、１２２Ｙ及び１２２Ｋを用いて用紙Ｐに形成された画像を光学的に読み取り、その読取画像を示す画像データを生成する装置である。スキャナ１３２は、用紙Ｐ上に印刷された画像を撮像して電気信号に変換する撮像デバイスを含む。撮像デバイスとしてカラーＣＣＤ（charge coupled device）リニアイメージセンサを用いることができる。なお、カラーＣＣＤリニアイメージセンサに代えて、カラーＣＭＯＳ（complementary metal oxide semiconductor）リニアイメージセンサを用いることもできる。 The scanner 132 is a device that optically reads an image formed on the sheet P using the inkjet heads 122C, 122M, 122Y, and 122K, and generates image data indicating the read image. The scanner 132 includes an imaging device that captures an image printed on the sheet P and converts the image into an electrical signal. A color charge coupled device (CCD) linear image sensor can be used as an imaging device. A color CMOS (complementary metal oxide semiconductor) linear image sensor may be used instead of the color CCD linear image sensor.
　スキャナ１３２は、撮像デバイスの他、読み取り対象を照明する照明光学系及び撮像デバイスから得られる信号を処理してデジタル画像データを生成する信号処理回路を含んでもよい。 The scanner 132 may include, in addition to an imaging device, an illumination optical system that illuminates an object to be read, and a signal processing circuit that processes a signal obtained from the imaging device to generate digital image data.
　〔乾燥部〕
　乾燥部１４０は、用紙Ｐのインクを乾燥させる。乾燥部１４０は、用紙Ｐの搬送方向に対して撮像部１３０の下流側に配置される。 [Drying part]
The drying unit 140 dries the ink of the paper P. The drying unit 140 is disposed downstream of the imaging unit 130 in the transport direction of the sheet P.
　乾燥部１４０は、ヒータ１４２を備えている。ヒータ１４２としては、例えば、ハロゲンヒータ及び赤外線ヒータのうち少なくとも一方が使用される。ヒータ１４２は、用紙Ｐの印刷面を加熱して、用紙Ｐのインクを乾燥させる。乾燥部１４０は、ファン又はブロアなどの送風手段を含んでいてもよい。 The drying unit 140 includes a heater 142. As the heater 142, for example, at least one of a halogen heater and an infrared heater is used. The heater 142 heats the printing surface of the sheet P to dry the ink of the sheet P. The drying unit 140 may include blowing means such as a fan or a blower.
　〔選別部〕
　選別部１５０は、搬送ベルト１１６によって搬送される用紙Ｐに関する良否判定に応じて、印刷物を選別する。選別部１５０は、用紙Ｐの搬送方向に対して乾燥部１４０の下流側に配置される。選別部１５０は、スタンパ１５２を備えている。 [Sorting unit]
The sorting unit 150 sorts the printed matter according to the quality determination regarding the sheet P conveyed by the conveyance belt 116. The sorting unit 150 is disposed downstream of the drying unit 140 in the transport direction of the sheet P. The sorting unit 150 includes a stamper 152.
　スタンパ１５２は、搬送ベルト１１６によって搬送される用紙Ｐに関する良否判定に応じて、不良品印刷物と判定された用紙Ｐの先端エッジにインクを付着させるスタンプ処理を行う。 The stamper 152 performs a stamping process to attach the ink to the leading edge of the sheet P determined to be a defective printed matter according to the quality determination regarding the sheet P transported by the transport belt 116.
　〔排紙部〕
　排紙部１６０は、画像が形成され、乾燥された用紙Ｐ（印刷物）を回収する。排紙部１６０は、用紙Ｐの搬送方向に対して選別部１５０の下流側であって、搬送部１１０の搬送経路の終点に配置される。排紙部１６０は、排紙台１６２を備えている。 [Paper output unit]
The paper discharge unit 160 collects the sheet P (printed matter) on which the image is formed and dried. The paper discharge unit 160 is disposed downstream of the sorting unit 150 in the conveyance direction of the sheet P and at the end point of the conveyance path of the conveyance unit 110. The paper discharge unit 160 includes a paper discharge stand 162.
　排紙台１６２は、搬送ベルト１１６によって搬送された用紙Ｐを積み重ねて回収する。排紙台１６２には、不図示の前用紙当て、後用紙当て、横用紙当てが備えられており、用紙Ｐを整然と積み重ねる。 The paper discharge tray 162 stacks and collects the sheets P conveyed by the conveyance belt 116. The paper output tray 162 is provided with a front sheet contact, a rear sheet contact, and a horizontal sheet contact (not shown), and the sheets P are stacked in order.
　また、排紙台１６２は、不図示の昇降装置によって昇降可能に設けられる。昇降装置は、排紙台１６２に積み重ねられる用紙Ｐの増減に連動して駆動が制御される。これにより、排紙台１６２に積み重ねられた用紙Ｐのうち最上位に位置する用紙Ｐが常に一定の高さとなる。 Further, the paper discharge tray 162 is provided so as to be able to move up and down by a lifting device (not shown). Driving of the lifting and lowering device is controlled in conjunction with increase and decrease of the sheets P stacked on the sheet discharge tray 162. As a result, among the sheets P stacked on the sheet discharge tray 162, the top sheet P is always at a constant height.
　〔インクジェット印刷装置の制御系〕
　図１０は、インクジェット印刷装置１００の内部構成を示すブロック図である。インクジェット印刷装置１００は、前述の画像欠陥検出装置１２、搬送部１１０、印刷部１２０、撮像部１３０、乾燥部１４０、選別部１５０、及び排紙部１６０の他、ユーザインターフェース１７０、記憶部１７２、統括制御部１７４、搬送制御部１７６、印刷制御部１７８、撮像制御部１８０、乾燥制御部１８２、選別制御部１８４、及び排紙制御部１８６を備えている。 [Control system of inkjet printing apparatus]
FIG. 10 is a block diagram showing the internal configuration of the inkjet printing apparatus 100. As shown in FIG. The inkjet printing apparatus 100 includes a user interface 170, a storage unit 172, in addition to the image defect detection device 12, the transport unit 110, the printing unit 120, the imaging unit 130, the drying unit 140, the sorting unit 150, and the paper discharge unit 160. The general control unit 174, the conveyance control unit 176, the print control unit 178, the imaging control unit 180, the drying control unit 182, the sorting control unit 184, and the paper discharge control unit 186 are provided.
　ユーザインターフェース１７０は、ユーザがインクジェット印刷装置１００を操作するための不図示の入力部及び不図示の表示部を備えている。入力部は、例えばユーザからの入力を受け付ける操作パネルである。表示部は、例えば画像データ及び各種の情報を表示するディスプレイである。ユーザは、ユーザインターフェース１７０を操作することで、インクジェット印刷装置１００に所望の画像を印刷させることができる。 The user interface 170 includes an input unit (not shown) and a display unit (not shown) for the user to operate the inkjet printing apparatus 100. The input unit is, for example, an operation panel that receives an input from a user. The display unit is, for example, a display that displays image data and various information. The user can operate the user interface 170 to cause the inkjet printing apparatus 100 to print a desired image.
　記憶部１７２は、インクジェット印刷装置１００を制御するためのプログラム及びプログラムの実行に必要な情報を記憶する。記憶部１７２は、不図示のハードディスク、又は各種半導体メモリ等の非一時的記録媒体により構成される。記憶部１７２は、基準画像Ｄ１等を一時的に記憶する不図示のＲＡＭ（Random Access Memory）等の揮発性のメモリを備えていてもよい。 The storage unit 172 stores a program for controlling the inkjet printing apparatus 100 and information necessary for executing the program. The storage unit 172 is configured by a non-temporary recording medium such as a hard disk (not shown) or various semiconductor memories. The storage unit 172 may include volatile memory such as a random access memory (RAM) (not shown) that temporarily stores the reference image D1 and the like.
　統括制御部１７４は、記憶部１７２に記憶されたプログラムに従って各種の処理を行い、インクジェット印刷装置１００の全体の動作を統括制御する。 The overall control unit 174 performs various processes in accordance with the program stored in the storage unit 172, and generally controls the overall operation of the inkjet printing apparatus 100.
　搬送制御部１７６は、搬送部１１０の不図示のモータを制御することで、搬送部１１０によって用紙Ｐを搬送方向に搬送させる。これにより、不図示の給紙部から供給された用紙Ｐは、印刷部１２０、撮像部１３０、乾燥部１４０、及び選別部１５０と対向する位置を通過し、最後に排紙部１６０に排紙される。 The conveyance control unit 176 causes the conveyance unit 110 to convey the sheet P in the conveyance direction by controlling a motor (not shown) of the conveyance unit 110. Thus, the sheet P supplied from the sheet feeding unit (not shown) passes through the position facing the printing unit 120, the imaging unit 130, the drying unit 140, and the sorting unit 150, and is finally discharged to the paper delivery unit 160. Be done.
　印刷制御部１７８は、インクジェットヘッド１２２Ｃ、１２２Ｍ、１２２Ｙ、及び１２２Ｋによるインクの吐出を制御する。印刷制御部１７８は、インクジェットヘッド１２２Ｃ、１２２Ｍ、１２２Ｙ、及び１２２Ｋによって、それぞれのノズル面と対向する位置を用紙Ｐが通過するタイミングにおいて、シアン、マゼンタ、イエロー、及びクロのインク滴を用紙Ｐに向けて吐出させる。これにより、用紙Ｐの印刷面にカラー画像が形成され、用紙Ｐは「印刷物」となる。 The print control unit 178 controls the ejection of the ink by the inkjet heads 122C, 122M, 122Y, and 122K. The print control unit 178 applies ink droplets of cyan, magenta, yellow, and black onto the sheet P at timing when the sheet P passes the positions facing the respective nozzle surfaces by the inkjet heads 122C, 122M, 122Y, and 122K. Direct the discharge. Thereby, a color image is formed on the printing surface of the sheet P, and the sheet P becomes a "printed material".
　撮像制御部１８０は、スキャナ１３２による撮像を制御することで、撮像部１３０によって用紙Ｐ（印刷物）の画像を読み取らせる。撮像制御部１８０は、スキャナ１３２によって、スキャナ１３２と対向する位置を用紙Ｐが通過するタイミングにおいて、用紙Ｐに形成された画像を読み取らせる。これにより、検査画像を取得する。 The imaging control unit 180 causes the imaging unit 130 to read the image of the sheet P (printed matter) by controlling the imaging by the scanner 132. The imaging control unit 180 causes the scanner 132 to read the image formed on the sheet P at the timing when the sheet P passes the position facing the scanner 132. Thereby, an examination image is acquired.
　乾燥制御部１８２は、ヒータ１４２による加熱を制御することで、乾燥部１４０によって用紙Ｐを乾燥させる。乾燥制御部１８２は、ヒータ１４２によって、ヒータ１４２と対向する位置を用紙Ｐが通過する際に用紙Ｐを加熱させる。 The drying control unit 182 controls the heating by the heater 142 to dry the sheet P by the drying unit 140. The drying control unit 182 causes the heater 142 to heat the sheet P when the sheet P passes a position facing the heater 142.
　選別制御部１８４は、スタンパ１５２によるスタンプ処理を制御することで、選別部１５０によって用紙Ｐを選別させる。選別制御部１８４は、スタンパ１５２と対向する位置を通過する用紙Ｐが不良品印刷物と判定された用紙Ｐである場合は、スタンパ１５２によってスタンプ処理を行う。 The sorting control unit 184 causes the sorting unit 150 to sort the sheets P by controlling the stamp processing by the stamper 152. When the sheet P passing through the position facing the stamper 152 is the sheet P judged to be a defective product printed matter, the selection control unit 184 performs the stamp processing by the stamper 152.
　排紙制御部１８６は、排紙台１６２による用紙Ｐの積載を制御する。用紙Ｐは、排紙台１６２に排紙され、積み重ねられる。不良品印刷物の用紙Ｐには先端エッジにインクが付着している。このため、ユーザは、排紙台１６２に積載された用紙Ｐの中から不良品印刷物を特定することができる。 The discharge control unit 186 controls stacking of the sheets P by the discharge tray 162. The sheet P is discharged to the discharge tray 162 and stacked. Ink adheres to the leading edge of the sheet P of the defective product printed matter. For this reason, the user can specify a defective product print out of the sheets P stacked on the sheet discharge tray 162.
　〔印刷方法〕
　インクジェット印刷装置１００における印刷方法について説明する。図１１は、印刷方法の処理と各処理に用いるデータについて示したフローチャートである。なお、図４に示すフローチャートと共通する部分には同一の符号を付し、その詳細な説明は省略する。本実施形態では、インクジェット印刷装置１００は、１枚の用紙Ｐに１ページ分の画像を印刷する。 [Printing method]
The printing method in the inkjet printing apparatus 100 will be described. FIG. 11 is a flowchart showing processing of the printing method and data used for each processing. The parts common to the flowchart shown in FIG. 4 are denoted by the same reference numerals, and the detailed description thereof is omitted. In the present embodiment, the inkjet printing apparatus 100 prints an image of one page on one sheet of paper P.
　最初に、ステップＳ４１（画像取得工程の一例）において、画像欠陥検出装置１２の画像取得部２０（基準画像取得部の一例）は、ユーザインターフェース１７０又は記憶部１７２から、基準画像Ｄ１を取得する。異なる複数のページを印刷する印刷ジョブであれば、全てのページの基準画像Ｄ１を取得する。 First, in step S41 (an example of an image acquisition step), the image acquisition unit 20 (an example of a reference image acquisition unit) of the image defect detection device 12 acquires a reference image D1 from the user interface 170 or the storage unit 172. In the case of a print job for printing different pages, reference images D1 of all pages are acquired.
　次に、ステップＳ４２では、搬送部１１０において用紙Ｐの搬送を開始する。また、ユーザインターフェース１７０又は記憶部１７２から取得した印刷元データに基づいて、印刷部１２０において用紙Ｐに印刷を行う。 Next, in step S42, conveyance of the sheet P is started in the conveyance unit 110. The printing unit 120 prints on the sheet P based on the print source data acquired from the user interface 170 or the storage unit 172.
　続いて、ステップＳ４３（画像取得工程の一例）では、撮像部１３０において用紙Ｐの画像を撮像し、検査画像Ｄ２を取得する。 Subsequently, in step S43 (an example of the image acquisition process), the image of the sheet P is imaged by the imaging unit 130 to acquire the inspection image D2.
　基準画像Ｄ１及び検査画像Ｄ２を取得すると、ステップＳ２に移行する。ステップＳ２では、画像欠陥検出装置１２の第１検出処理部２２において第１検出処理を行い、検査画像Ｄ２の欠陥候補位置Ｄ３及びその強度を検出する。 When the reference image D1 and the inspection image D2 are acquired, the process proceeds to step S2. In step S2, the first detection processing unit 22 of the image defect detection device 12 performs a first detection process to detect the defect candidate position D3 of the inspection image D2 and the intensity thereof.
　次に、ステップＳ１１において、画像欠陥検出装置１２の選択部２４は、複数の欠陥候補位置Ｄ３の中から上限値Ｄ４以下の数の選択欠陥候補位置Ｄ５を選択する。ここでは、選択部２４は、強度が弱い欠陥候補位置Ｄ３を優先して選択する。また、選択部２４は、複数の欠陥候補位置Ｄ３のうち選択欠陥候補位置Ｄ５以外の欠陥候補位置を非選択欠陥候補位置Ｄ１１とする。非選択欠陥候補位置Ｄ１１は、画像欠陥検出装置１２の判断部２８に入力される。 Next, in step S11, the selection unit 24 of the image defect detection device 12 selects the number of selected defect candidate positions D5 equal to or less than the upper limit value D4 from among the plurality of defect candidate positions D3. Here, the selection unit 24 preferentially selects the defect candidate position D3 whose strength is weak. Further, the selection unit 24 sets a defect candidate position other than the selected defect candidate position D5 among the plurality of defect candidate positions D3 as a non-selected defect candidate position D11. The non-selected defect candidate position D11 is input to the determination unit 28 of the image defect detection device 12.
　続くステップＳ１２では、判断部２８は、入力された非選択欠陥候補位置Ｄ１１について、画像欠陥が存在すると判断する。 In the subsequent step S12, the determination unit 28 determines that an image defect exists at the input non-selected defect candidate position D11.
　次に、ステップＳ４において、画像欠陥検出装置１２の第２検出処理部２６は、選択欠陥候補位置Ｄ５に対して画像欠陥Ｄ６を検出する。 Next, in step S4, the second detection processing unit 26 of the image defect detection device 12 detects the image defect D6 at the selected defect candidate position D5.
　さらに、ステップＳ１３において、画像欠陥検出装置１２の良否判定部３０は、判断部２８の判断結果Ｄ１２による画像欠陥Ｄ６と第２検出処理部２６の検出結果である画像欠陥Ｄ６に基づいて、印刷物の良否判定を行い、良否判定結果Ｄ１３を出力する。良否判定結果Ｄ１３は、選別制御部１８４に入力される。 Furthermore, in step S13, the quality determination unit 30 of the image defect detection device 12 determines the printed matter based on the image defect D6 according to the determination result D12 of the determination unit 28 and the image defect D6 which is the detection result of the second detection processing unit 26. A pass / fail judgment is made, and a pass / fail judgment result D13 is output. The quality determination result D13 is input to the selection control unit 184.
　ステップＳ４４では、選別制御部１８４は、良否判定結果Ｄ１３に基づいて、スタンパ１５２と対向する位置を通過する用紙Ｐが良品印刷物であるか否かを判定する。良品印刷物でない場合、即ち不良品印刷物である場合は、ステップＳ４５へ移行する。また、良品印刷物である場合は、ステップＳ４６へ移行する。 In step S44, the selection control unit 184 determines whether or not the sheet P passing the position facing the stamper 152 is a non-defective printed material based on the quality determination result D13. If it is not a non-defective print, that is, if it is a non-defective print, the process proceeds to step S45. If the print is non-defective, the process proceeds to step S46.
　ステップＳ４５では、スタンパ１５２によるスタンプ処理を行う。これにより、不良品印刷物の特定が容易になる。スタンプ処理を行った後、ステップＳ４６へ移行する。 In step S45, the stamp processing by the stamper 152 is performed. This facilitates identification of defective printed matter. After performing the stamp processing, the process proceeds to step S46.
　ステップＳ４６では、統括制御部１７４は、印刷ジョブの全てのページの印刷が終了したか否かを判定する。印刷を終了していないページが存在する場合は、Ｓ４２に移行し、次の用紙Ｐの印刷を行う。 In step S46, the general control unit 174 determines whether printing of all pages of the print job is completed. If there is a page for which printing has not been completed, the process proceeds to S42, and printing of the next sheet P is performed.
　全てのページの印刷が終了した場合は、本フローチャートの処理を終了する。 If printing of all pages is completed, the processing of this flowchart ends.
　インクジェット印刷装置１００のように、オンラインで画像欠陥検出処理を行う場合、画像欠陥検出処理を１ページ分の印刷時間内（用紙Ｐの１枚あたりの印刷時間内）に終了させる必要がある。インクジェット印刷装置１００によれば、検出時間の目標時間である１ページ分の印刷時間と第１検出処理部２２及び第２検出処理部２６の処理能力とに応じて、第２検出処理を行う欠陥候補位置の数の上限値を設定することができるので、画像欠陥検出処理が目標時間内に収まらなくなるということは発生しない。したがって、画像欠陥検出処理を適切に行いつつ、目標時間内に終了させることができる。 When the image defect detection process is performed online as in the inkjet printing apparatus 100, the image defect detection process needs to be completed within the print time of one page (within the print time per sheet of paper P). According to the inkjet printing apparatus 100, the defect for performing the second detection process according to the printing time for one page, which is the target time of the detection time, and the processing capability of the first detection processing unit 22 and the second detection processing unit 26. Since the upper limit of the number of candidate positions can be set, it does not occur that the image defect detection process does not fall within the target time. Therefore, it is possible to complete the process within the target time while properly performing the image defect detection process.
　ここでは、画像欠陥検出装置１２を適用したインクジェット印刷装置について説明したが、画像欠陥検出装置１４及び１６を適用した場合も同様である。また、画像欠陥検出装置１０を適用する場合は、検出された画像欠陥Ｄ６に基づいて統括制御部１７４等が印刷物の良否判定を行えばよい。 Here, the inkjet printing apparatus to which the image defect detection apparatus 12 is applied has been described, but the same applies to the case where the image defect detection apparatuses 14 and 16 are applied. When the image defect detection apparatus 10 is applied, the general control unit 174 or the like may determine the quality of the printed matter based on the detected image defect D6.
　本実施形態では、基準画像Ｄ１を取得してから印刷を開始しているが、基準画像Ｄ１は必ずしも印刷開始前に取得しておく必要はない。例えば、１ページ目の撮像画像を基準画像Ｄ１とし、２ページ目から画像欠陥検出処理を開始してもよい。即ち、基準画像Ｄ１は、第１検出処理を行うタイミングまでに取得すればよい。 In the present embodiment, printing is started after acquiring the reference image D1, but the reference image D1 does not necessarily have to be acquired before the start of printing. For example, the captured image of the first page may be set as the reference image D1, and the image defect detection process may be started from the second page. That is, the reference image D1 may be acquired before the timing at which the first detection process is performed.
　また、各工程を必ずしも逐次的に行うことに限定されない。例えば、あるページの画像欠陥処理を実施している間に次のページの検査画像読取を実施し、さらにその次のページの印刷を実施する等、各工程を並列的に実施してもよい。 Moreover, it is not limited to necessarily performing each process one by one. For example, each process may be performed in parallel, such as performing inspection image reading of the next page while performing image defect processing of a certain page, and further performing printing of the next page.
　ここでは、「ページ」という用語を使用しているが、これは枚葉の用紙Ｐを印刷する印刷装置に限定する用語ではない。ロール紙に印刷する印刷装置であっても、ページに分割して印刷を行い、それぞれ検査することが可能である。 Here, the term "page" is used, but this is not a term that is limited to a printing apparatus that prints the sheet of paper P. Even in the case of a printing apparatus that prints on roll paper, it is possible to print by dividing into pages and inspect each of them.
　＜第２検出処理を経過時間ではなく欠陥候補位置の上限値で制限する利点＞
　第２検出処理を行う欠陥候補位置の数に上限値を設けるのではなく、第２検出処理における実施時間を計測しておき、ある時間に達したら第２検出処理を中断するという態様でも同様の効果を得ることが可能である。しかしながら、上限値を設けることで次のような利点がある。 <The advantage of limiting the second detection process not by the elapsed time but by the upper limit value of the defect candidate position>
Instead of providing an upper limit to the number of defect candidate positions for which the second detection process is performed, the execution time in the second detection process is measured, and the second detection process is interrupted when it reaches a certain time. It is possible to obtain an effect. However, setting the upper limit value has the following advantages.
　本発明のような画像欠陥検出処理は、前述のように、並列演算器を用いることが可能である。例えば、並列演算器として複数のＧＰＵを用いた場合、演算コア数、単一演算コアあたりの処理能力、及びトータルのメモリ量等で同時に処理することができる欠陥候補位置の数が決まる。このため、ＧＰＵの処理能力に合わせて上限値を設定することが容易である。したがって、少ない労力で目標時間内での画像欠陥検出処理の終了を保証することができる。 The image defect detection processing as in the present invention can use a parallel computing unit as described above. For example, when a plurality of GPUs are used as parallel computing units, the number of defect candidate positions that can be processed simultaneously is determined by the number of operation cores, the processing capacity per single operation core, and the total memory amount. For this reason, it is easy to set the upper limit value in accordance with the processing capacity of the GPU. Therefore, it is possible to guarantee the end of the image defect detection process within the target time with little effort.
　また、ＧＰＵのタスク割付は、割付を行う際のリソース残量に応じて自動で行われることが多い。このため、処理を開始した後に途中で中断することが難しい。途中で中断するためには、処理の途中に同期処理（各演算コアそれぞれが所定の処理を完了するまで互いに待つ動作）を設けて同期のタイミングで経過時間を確認し、必要であれば中断する、というような動作をさせる必要性が生じる。しかしながら、同期処理を行うことで演算コア間に待ち時間が発生するため、同期処理を行わない場合に比べ、ある時間内に処理できる欠陥候補位置の数が減少してしまう。また、同期処理を行うため設計が複雑になるといった問題も発生する。上限値を用いることでこのような同期処理を回避することが可能である。 Also, task assignment of the GPU is often performed automatically according to the remaining amount of resources at the time of assignment. For this reason, it is difficult to interrupt the process after starting the process. In order to interrupt on the way, provide synchronous processing (operation where each operation core waits for each processing core to complete each other) in the middle of processing, check the elapsed time at the timing of synchronization, and interrupt if necessary. There is a need to have an action such as However, since the execution of the synchronization processing causes a waiting time between the operation cores, the number of defect candidate positions that can be processed within a certain time is reduced compared to the case where the synchronization processing is not performed. In addition, there is a problem that the design becomes complicated due to the synchronization process. It is possible to avoid such synchronization processing by using the upper limit value.
　＜その他＞
　上記の画像欠陥検出方法は、各工程をコンピュータに実現させるためのプログラムとして構成し、このプログラムを記憶したＣＤ－ＲＯＭ（Compact Disk-Read Only Memory）等の非一時的な記録媒体を構成することも可能である。 <Others>
The image defect detection method described above is configured as a program for causing a computer to realize each process, and to configure a non-temporary recording medium such as a CD-ROM (Compact Disk-Read Only Memory) storing the program. Is also possible.
　ここまで説明した実施形態において、例えば、画像欠陥検出装置１０、１２、１４、及び１６の各種の処理を実行する処理部（processing unit）のハードウェア的な構造は、次に示すような各種のプロセッサ（processor）である。各種のプロセッサには、ソフトウェア（プログラム）を実行して各種の処理部として機能する汎用的なプロセッサであるＣＰＵ（Central Processing Unit）、画像処理に特化したプロセッサであるＧＰＵ（Graphics Processing Unit）、ＦＰＧＡ（Field Programmable Gate Array）等の製造後に回路構成を変更可能なプロセッサであるプログラマブルロジックデバイス（Programmable Logic Device：ＰＬＤ）、ＡＳＩＣ（Application Specific Integrated Circuit）等の特定の処理を実行させるために専用に設計された回路構成を有するプロセッサである専用電気回路等が含まれる。 In the embodiment described above, for example, the hardware-like structure of the processing unit that executes various processes of the image defect detection devices 10, 12, 14, and 16 is various as shown below. It is a processor. Various processors include a CPU (Central Processing Unit), which is a general-purpose processor that executes software (programs) and functions as various processing units, and a GPU (Graphics Processing Unit), which is a processor specialized for image processing, Dedicated to execute specific processing such as Programmable Logic Device (PLD) or ASIC (Application Specific Integrated Circuit), which is a processor whose circuit configuration can be changed after manufacturing an FPGA (Field Programmable Gate Array) or the like It includes a dedicated electric circuit or the like which is a processor having a designed circuit configuration.
　１つの処理部は、これら各種のプロセッサのうちの１つで構成されていてもよいし、同種又は異種の２つ以上のプロセッサ（例えば、複数のＦＰＧＡ、あるいはＣＰＵとＦＰＧＡの組み合わせ、又はＣＰＵとＧＰＵの組み合わせ）で構成されてもよい。また、複数の処理部を１つのプロセッサで構成してもよい。複数の処理部を１つのプロセッサで構成する例としては、第１に、サーバ及びクライアント等のコンピュータに代表されるように、１つ以上のＣＰＵとソフトウェアの組合せで１つのプロセッサを構成し、このプロセッサが複数の処理部として機能する形態がある。第２に、システムオンチップ（System On Chip：ＳｏＣ）等に代表されるように、複数の処理部を含むシステム全体の機能を１つのＩＣ（Integrated Circuit）チップで実現するプロセッサを使用する形態がある。このように、各種の処理部は、ハードウェア的な構造として、各種のプロセッサを１つ以上用いて構成される。 One processing unit may be configured of one of these various types of processors, or two or more processors of the same or different types (for example, a plurality of FPGAs, a combination of a CPU and an FPGA, or a CPU and (A combination of GPUs). In addition, a plurality of processing units may be configured by one processor. As an example in which a plurality of processing units are configured by one processor, first, one processor or more is configured by a combination of one or more CPUs and software as represented by computers such as servers and clients. There is a form in which a processor functions as a plurality of processing units. Second, as typified by a system on chip (SoC) or the like, there is a form using a processor that realizes the functions of the entire system including a plurality of processing units in one integrated circuit (IC) chip. is there. As described above, the various processing units are configured using one or more of various processors as a hardware structure.
　さらに、これらの各種のプロセッサのハードウェア的な構造は、より具体的には、半導体素子等の回路素子を組み合わせた電気回路（circuitry）である。 Furthermore, the hardware-like structure of these various processors is more specifically an electric circuit (circuitry) combining circuit elements such as semiconductor elements.
　本発明の技術的範囲は、上記の実施形態に記載の範囲には限定されない。各実施形態における構成等は、本発明の趣旨を逸脱しない範囲で、各実施形態間で適宜組み合わせることができる。 The technical scope of the present invention is not limited to the scope described in the above embodiment. The configuration and the like in each embodiment can be appropriately combined among the embodiments without departing from the spirit of the present invention.
１０　画像欠陥検出装置
１２　画像欠陥検出装置
１４　画像欠陥検出装置
１６　画像欠陥検出装置
２０　画像取得部
２２　第１検出処理部
２４　選択部
２６　第２検出処理部
２８　判断部
３０　良否判定部
３２　格納部
３４　通知部
３６　設定部
１００　インクジェット印刷装置
１１０　搬送部
１１２　上流側プーリ
１１４　下流側プーリ
１１６　搬送ベルト
１２０　印刷部
１２２Ｃ　インクジェットヘッド
１２２Ｋ　インクジェットヘッド
１２２Ｍ　インクジェットヘッド
１２２Ｙ　インクジェットヘッド
１３０　撮像部
１３２　スキャナ
１４０　乾燥部
１４２　ヒータ
１５０　選別部
１５２　スタンパ
１６０　排紙部
１６２　排紙台
１７０　ユーザインターフェース
１７２　記憶部
１７４　統括制御部
１７６　搬送制御部
１７８　印刷制御部
１８０　撮像制御部
１８２　乾燥制御部
１８４　選別制御部
１８６　排紙制御部
Ｄ１　基準画像
Ｄ２　検査画像
Ｄ３　欠陥候補位置
Ｄ４　上限値
Ｄ５　選択欠陥候補位置
Ｄ６　画像欠陥
Ｄ１１　非選択欠陥候補位置
Ｄ１２　判断結果
Ｄ１３　良否判定結果
Ｐ　用紙
Ｓ１～Ｓ３４　画像欠陥検出方法の工程
Ｓ１～Ｓ４６　印刷方法の工程 10 Image defect detection device 12 Image defect detection device 14 Image defect detection device 16 Image defect detection device 20 Image acquisition unit 22 First detection processing unit 24 Selection unit 26 Second detection processing unit 28 Judgment unit 30 Quality judgment unit 32 Storage unit 34 Notification unit 36 Setting unit 100 Inkjet printing apparatus 110 Conveying unit 112 Upstream pulley 114 Downstream pulley 116 Conveying belt 120 Printing unit 122C Inkjet head 122K Inkjet head 122M Inkjet head 122Y Inkjet head 130 Imaging unit 132 Scanner 140 Drying unit 142 Heater 150 Selection Part 152 Stamper 160 Discharge part 162 Sheet ejection tray 170 User interface 172 Storage part 174 General control part 176 Transport control part 178 Print control part 180 Imaging control part 182 Drying control part 184 Sorting Control section 186 discharge control unit D1 reference image D2 inspection image D3 defect candidate position D4 upper limit value D5 selected defect candidate position D6 image defect D11 non-selected defect candidate position D12 judgment result D13 good / not good judgment result P paper S1 to S34 image defect detection method Steps S1-S46 of the printing method

Claims (23)

  1. 　印刷物を撮像した検査画像と前記印刷物の画像欠陥を検出する基準となる基準画像とを取得する画像取得部と、
    　第１検出処理により、前記検査画像から画像欠陥の候補の位置を示す欠陥候補位置を検出する第１検出処理部と、
    　前記検査画像と前記基準画像とを比較する第２検出処理により、前記欠陥候補位置の画像欠陥を検出する第２検出処理部と、
    　前記検出された欠陥候補位置のうち前記第２検出処理を行う欠陥候補位置を選択する選択部と、
    　を備え、
    　前記選択部は、上限値以下の数の前記欠陥候補位置を選択し、
    　前記第２検出処理部は、前記選択された欠陥候補位置についてのみ前記第２検出処理を行う画像欠陥検出装置。     An image acquisition unit for acquiring an inspection image obtained by imaging a printed matter and a reference image serving as a reference for detecting an image defect of the printed matter;
    A first detection processing unit configured to detect a defect candidate position indicating a position of an image defect candidate from the inspection image by a first detection process;
    A second detection processing unit configured to detect an image defect of the defect candidate position by a second detection process of comparing the inspection image and the reference image;
    A selection unit that selects a defect candidate position to be subjected to the second detection process among the detected defect candidate positions;
    Equipped with
    The selection unit selects the number of defect candidate positions equal to or less than an upper limit value,
    The image defect detection device according to claim 1, wherein the second detection processing unit performs the second detection process only for the selected defect candidate position.
  2. 　前記第１検出処理部は、前記欠陥候補位置の強度を検出し、
    　前記選択部は、前記検出した強度が弱い欠陥候補位置を優先して選択し、
    　前記欠陥候補位置のうち前記選択部において選択されない欠陥候補位置については画像欠陥が存在すると判断する判断部を備えた請求項１に記載の画像欠陥検出装置。     The first detection processing unit detects the strength of the defect candidate position,
    The selection unit preferentially selects a defect candidate position where the detected intensity is weak,
    The image defect detection apparatus according to claim 1, further comprising a determination unit configured to determine that an image defect is present at a defect candidate position not selected by the selection unit among the defect candidate positions.
  3. 　前記第１検出処理部は、前記欠陥候補位置の強度を検出し、
    　前記選択部は、前記検出した強度が強い欠陥候補位置を優先して選択する請求項１に記載の画像欠陥検出装置。     The first detection processing unit detects the strength of the defect candidate position,
    The image defect detection device according to claim 1, wherein the selection unit preferentially selects a defect candidate position where the detected intensity is high.
  4. 　前記欠陥候補位置のうち前記選択部において選択されない欠陥候補位置については画像欠陥が存在しないと判断する判断部を備えた請求項３に記載の画像欠陥検出装置。 The image defect detection apparatus according to claim 3, further comprising: a determination unit that determines that no image defect exists at a defect candidate position not selected by the selection unit among the defect candidate positions.
  5. 　前記欠陥候補位置を検出した順に格納する格納部を備え、
    　前記選択部は、前記欠陥候補位置を前記格納部に格納された正順又は逆順に選択する請求項１に記載の画像欠陥検出装置。     A storage unit for storing the defect candidate positions in the order of detection;
    The image defect detection device according to claim 1, wherein the selection unit selects the defect candidate positions in the normal order or the reverse order stored in the storage unit.
  6. 　前記欠陥候補位置の数が前記上限値を超えた場合にユーザに通知する通知部を備えた請求項１から５のいずれか１項に記載の画像欠陥検出装置。 The image defect detection device according to any one of claims 1 to 5, further comprising: a notification unit configured to notify the user when the number of defect candidate positions exceeds the upper limit value.
  7. 　前記通知部による通知の有無を設定する設定部を備えた請求項６に記載の画像欠陥検出装置。 The image defect detection apparatus according to claim 6, further comprising a setting unit configured to set presence / absence of notification by the notification unit.
  8. 　画像全体に対する前記第２検出処理の演算量は、前記画像全体に対する前記第１検出処理の演算量よりも大きい請求項１から７のいずれか１項に記載の画像欠陥検出装置。 The image defect detection device according to any one of claims 1 to 7, wherein the operation amount of the second detection process for the entire image is larger than the operation amount of the first detection process for the entire image.
  9. 　画像の単一領域に対する前記第２検出処理の演算量は、前記画像の単一領域に対する前記第１検出処理の演算量よりも大きい請求項１から８のいずれか１項に記載の画像欠陥検出装置。 The image defect detection according to any one of claims 1 to 8, wherein the operation amount of the second detection process for a single area of an image is larger than the operation amount of the first detection process for a single area of the image. apparatus.
  10. 　前記第１検出処理は、前記検査画像と前記基準画像とを使用する請求項１から９のいずれか１項に記載の画像欠陥検出装置。 The image defect detection device according to any one of claims 1 to 9, wherein the first detection process uses the inspection image and the reference image.
  11. 　前記第２検出処理は、前記欠陥候補位置を含む検査部分画像と前記検査部分画像に対応する基準部分画像を使用する請求項１から１０のいずれか１項に記載の画像欠陥検出装置。 The image defect detection device according to any one of claims 1 to 10, wherein the second detection process uses an inspection partial image including the defect candidate position and a reference partial image corresponding to the inspection partial image.
  12. 　前記第２検出処理は周波数解析処理を含む請求項１から１１のいずれか１項に記載の画像欠陥検出装置。 The image defect detection device according to any one of claims 1 to 11, wherein the second detection process includes a frequency analysis process.
  13. 　前記第２検出処理は多重解像度解析処理を含む請求項１から１２のいずれか１項に記載の画像欠陥検出装置。 The image defect detection device according to any one of claims 1 to 12, wherein the second detection process includes a multiresolution analysis process.
  14. 　前記第２検出処理は機械学習モデル又は深層学習モデルによる判定を含む請求項１から１３のいずれか１項に記載の画像欠陥検出装置。 The image defect detection device according to any one of claims 1 to 13, wherein the second detection process includes determination using a machine learning model or a deep learning model.
  15. 　前記機械学習モデル又は前記深層学習モデルを作成するための学習データには、人の官能評価実験によって得られた官能値又は官能値を加工したデータを含む請求項１４に記載の画像欠陥検出装置。 The image defect detection device according to claim 14, wherein the learning data for creating the machine learning model or the deep learning model includes data obtained by processing sensory values or sensory values obtained by human sensory evaluation experiments.
  16. 　前記第２検出処理部は、複数の演算を並列に処理する並列演算器を備える請求項１から１５のいずれか１項に記載の画像欠陥検出装置。 The image defect detection device according to any one of claims 1 to 15, wherein the second detection processing unit includes a parallel computing unit that processes a plurality of operations in parallel.
  17. 　前記並列演算器は複数のＣＰＵ（Central Processing Unit）を備える請求項１６に記載の画像欠陥検出装置。 The image defect detection device according to claim 16, wherein the parallel computing unit comprises a plurality of CPUs (central processing units).
  18. 　前記並列演算器は複数の基板を備える請求項１６又は１７に記載の画像欠陥検出装置。 The image defect detection device according to claim 16, wherein the parallel computing unit comprises a plurality of substrates.
  19. 　前記並列演算器は複数のＧＰＵ（Graphics Processing Unit）を備える請求項１６から１８のいずれか１項に記載の画像欠陥検出装置。 The image defect detection device according to any one of claims 16 to 18, wherein the parallel computing unit comprises a plurality of GPUs (Graphics Processing Units).
  20. 　印刷物を印刷する印刷部と、
    　前記印刷物を撮像して検査画像を取得する撮像部と、
    　前記印刷物の画像欠陥を検出する基準となる基準画像を取得する基準画像取得部と、
    　第１検出処理により、前記検査画像から画像欠陥の候補の位置を示す欠陥候補位置を検出する第１検出処理部と、
    　前記検査画像と前記基準画像とを比較する第２検出処理により、前記欠陥候補位置の画像欠陥を検出する第２検出処理部と、
    　前記検出された欠陥候補位置のうち前記第２検出処理を行う欠陥候補位置を選択する選択部と、
    　前記第２検出処理の検出結果に基づいて前記検査画像の良否の判定を行う良否判定部と、
    　を備え、
    　前記選択部は、上限値以下の数の前記欠陥候補位置を選択し、
    　前記第２検出処理部は、前記選択された欠陥候補位置についてのみ前記第２検出処理を行う印刷装置。     A printing unit for printing printed matter,
    An imaging unit for imaging the printed matter to acquire an inspection image;
    A reference image acquisition unit that acquires a reference image serving as a reference for detecting an image defect of the printed matter;
    A first detection processing unit configured to detect a defect candidate position indicating a position of an image defect candidate from the inspection image by a first detection process;
    A second detection processing unit configured to detect an image defect of the defect candidate position by a second detection process of comparing the inspection image and the reference image;
    A selection unit that selects a defect candidate position to be subjected to the second detection process among the detected defect candidate positions;
    A quality determination unit that determines the quality of the inspection image based on the detection result of the second detection process;
    Equipped with
    The selection unit selects the number of defect candidate positions equal to or less than an upper limit value,
    The second detection processing unit performs the second detection process only on the selected defect candidate position.
  21. 　印刷物を撮像した検査画像と前記印刷物の画像欠陥を検出する基準となる基準画像とを取得する画像取得工程と、
    　第１検出処理により、前記検査画像から画像欠陥の候補の位置を示す欠陥候補位置を検出する第１検出処理工程と、
    　前記検査画像と前記基準画像とを比較する第２検出処理により、前記欠陥候補位置の画像欠陥を検出する第２検出処理工程と、
    　前記検出された欠陥候補位置のうち前記第２検出処理を行う欠陥候補位置を選択する選択工程と、
    　を備え、
    　前記選択工程は、上限値以下の数の前記欠陥候補位置を選択し、
    　前記第２検出処理工程は、前記選択された欠陥候補位置についてのみ前記第２検出処理を行う画像欠陥検出方法。     An image acquisition step of acquiring an inspection image obtained by imaging a printed matter and a reference image serving as a reference for detecting an image defect of the printed matter;
    A first detection processing step of detecting a defect candidate position indicating a position of an image defect candidate from the inspection image by a first detection processing;
    A second detection processing step of detecting an image defect of the defect candidate position by a second detection processing of comparing the inspection image and the reference image;
    A selection step of selecting a defect candidate position to be subjected to the second detection process among the detected defect candidate positions;
    Equipped with
    The selection step selects the number of defect candidate positions equal to or less than an upper limit value,
    The image defect detection method, wherein the second detection processing step performs the second detection processing only for the selected defect candidate position.
  22. 　請求項２１に記載の画像欠陥検出方法をコンピュータに実行させるためのプログラム。 A program for causing a computer to execute the image defect detection method according to claim 21.
  23. 　請求項２２に記載のプログラムを記録したコンピュータ読み取り可能な非一時的記録媒体。 A computer readable non-transitory recording medium having the program according to claim 22 recorded thereon.
PCT/JP2018/044396 2018-01-17 2018-12-03 Printing apparatus, and device, method, and program for detecting image defects WO2019142517A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2018005699 2018-01-17
JP2018-005699 2018-01-17

Publications (1)

Publication Number Publication Date
WO2019142517A1 true WO2019142517A1 (en) 2019-07-25

Family

ID=67302175

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2018/044396 WO2019142517A1 (en) 2018-01-17 2018-12-03 Printing apparatus, and device, method, and program for detecting image defects

Country Status (1)

Country Link
WO (1) WO2019142517A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPWO2021054200A1 (en) * 2019-09-18 2021-03-25
JP2021047592A (en) * 2019-09-18 2021-03-25 富士フイルム株式会社 Defect inspection apparatus for printed matter, defect inspection method and program, and printing system

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000172844A (en) * 1998-12-03 2000-06-23 Toshiba Corp Printing inspection method and device therefor
JP2015174307A (en) * 2014-03-14 2015-10-05 株式会社リコー Image inspection device, image inspection system, and image inspection program
WO2016047377A1 (en) * 2014-09-22 2016-03-31 富士フイルム株式会社 Image-recording device, and image defect detection device and method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000172844A (en) * 1998-12-03 2000-06-23 Toshiba Corp Printing inspection method and device therefor
JP2015174307A (en) * 2014-03-14 2015-10-05 株式会社リコー Image inspection device, image inspection system, and image inspection program
WO2016047377A1 (en) * 2014-09-22 2016-03-31 富士フイルム株式会社 Image-recording device, and image defect detection device and method

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPWO2021054200A1 (en) * 2019-09-18 2021-03-25
JP2021047592A (en) * 2019-09-18 2021-03-25 富士フイルム株式会社 Defect inspection apparatus for printed matter, defect inspection method and program, and printing system
EP4032706A4 (en) * 2019-09-18 2022-10-19 FUJIFILM Corporation Printed matter inspecting device, printed matter inspecting method, program, and printing apparatus
JP7225068B2 (en) 2019-09-18 2023-02-20 富士フイルム株式会社 Defect inspection device for printed matter, defect inspection method and program, and printing system
JP7273169B2 (en) 2019-09-18 2023-05-12 富士フイルム株式会社 PRINTED MATERIAL INSPECTION DEVICE, PRINTED MATERIAL INSPECTION METHOD, PROGRAM AND PRINTING DEVICE

Similar Documents

Publication Publication Date Title
JP7110349B2 (en) Machine learning model generation device, method, program, inspection device and method, and printing device
JP6694362B2 (en) Image inspection method and apparatus, program, and image recording system
US9946944B2 (en) Image recording device, image defect detection device, and image defect detection method
JP6571575B2 (en) Line detection method and apparatus, printing apparatus and program
JP6433384B2 (en) Image processing apparatus and image processing method
US20220198638A1 (en) Printed matter inspection device, printed matter inspection method, program, and printing apparatus
JP5881576B2 (en) Image evaluation apparatus, image evaluation method, image evaluation system, and program
JP2020186938A (en) Inspection method, inspection device, program, and printing device
WO2019142517A1 (en) Printing apparatus, and device, method, and program for detecting image defects
JP6184386B2 (en) Image defect detection apparatus and method, image recording apparatus, and program
WO2021256389A1 (en) Defect inspection device, defect inspection method and program, printing device, and printed matter production method
WO2020012826A1 (en) Printing device, inspection device, inspection method and program
WO2019188911A1 (en) Image forming device and method, abnormal nozzle detection method and method of manufacturing printed matter
JP2017154468A (en) Image recording device and parameter setting method
JP6066395B2 (en) Image inspection apparatus and image inspection method
WO2020179742A1 (en) Inspection method, program, inspection device, and printing device
JP6293027B2 (en) Image defect detection apparatus and method, and program
JP7256139B2 (en) Defect inspection device for printed matter, defect inspection method and program, and printing system
JP6814883B2 (en) Image forming device and image forming method
JP7225068B2 (en) Defect inspection device for printed matter, defect inspection method and program, and printing system
JP7126477B2 (en) IMAGE INSPECTION METHOD AND APPARATUS, PROGRAM AND IMAGE RECORDING SYSTEM
JP7021144B2 (en) Printed matter inspection methods and programs, printed matter inspection equipment, printing systems
JP2018137693A (en) Image formation device and image formation device control method
JP6463257B2 (en) Image forming system and medium sorting method
JP2024070143A (en) Print inspection device, print inspection method, program, and printing system

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18901033

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18901033

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP