WO2023026706A1 - 画像処理装置及び画像処理方法 - Google Patents
画像処理装置及び画像処理方法 Download PDFInfo
- Publication number
- WO2023026706A1 WO2023026706A1 PCT/JP2022/027173 JP2022027173W WO2023026706A1 WO 2023026706 A1 WO2023026706 A1 WO 2023026706A1 JP 2022027173 W JP2022027173 W JP 2022027173W WO 2023026706 A1 WO2023026706 A1 WO 2023026706A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- control unit
- image processing
- processing device
- group
- Prior art date
Links
- 238000012545 processing Methods 0.000 title claims abstract description 114
- 238000003672 processing method Methods 0.000 title description 3
- 238000001514 detection method Methods 0.000 claims description 143
- 238000005070 sampling Methods 0.000 claims description 19
- 239000000284 extract Substances 0.000 claims description 18
- 230000000694 effects Effects 0.000 claims description 6
- 238000004891 communication Methods 0.000 abstract description 12
- 238000000034 method Methods 0.000 description 41
- 230000008569 process Effects 0.000 description 24
- 238000013179 statistical model Methods 0.000 description 18
- 238000004458 analytical method Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 7
- 238000000605 extraction Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 230000000052 comparative effect Effects 0.000 description 3
- 238000005286 illumination Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000010225 co-occurrence analysis Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
Definitions
- the present disclosure relates to an image processing device and an image processing method.
- Template matching is known in which the position of a detection target is detected from within an image of an arbitrary area using the image of the detection target as a template image.
- a detection target placed in an arbitrary area varies in appearance from a fixed template image due to various disturbance factors such as changes in the illuminance of sunlight, the position of a camera that captures an arbitrary area, and manual work on the detection target. can. Therefore, it has been proposed to perform robust template matching even against disturbance factors (see Non-Patent Documents 1 to 3).
- the image processing device is a memory for storing a template image group including a plurality of images to be detected; an acquisition unit that acquires an entire image including a matching target; a control unit that acquires a representative image based on a part of the template image group and performs similarity determination between at least part of the overall image and the representative image.
- the image processing device is a memory for storing a template image group including a plurality of images to be detected; an acquisition unit that acquires an entire image including a matching target; a control unit that acquires a representative image based on a part of the template image group and performs similarity determination between at least part of the overall image and the representative image; The control unit calculates a contribution degree of the part of the whole image used for the similarity judgment based on statistics of at least a part of the template image group, and uses the contribution degree for the similarity judgment.
- the image processing method is storing a template image group including a plurality of images to be detected; obtaining an entire image including a matching target; obtaining a representative image based on a portion of the template images; determining similarity between at least part of the overall image and the representative image.
- FIG. 1 is a diagram showing a configuration example of an object operation system including an image processing device according to the first embodiment.
- FIG. 2 is a diagram for explaining similarity determination between a representative image and a partial region of the overall image, which is performed by the control unit in FIG. 1 to detect the existing position of a detection target in the overall image;
- FIG. 2 is a diagram for explaining a method of creating a statistical model for the control unit in FIG. 1 to calculate a contribution;
- FIG. 10 is a diagram for explaining another method of creating a statistical model for the control unit in FIG. 1 to calculate the degree of contribution;
- FIG. 10 is a diagram for explaining a method of creating still another statistical model for the control unit in FIG. 1 to calculate the degree of contribution;
- FIG. 2 is a diagram for explaining similarity determination between a representative image and a partial region of the overall image, which is performed by the control unit in FIG. 1 to detect the existing position of a detection target in the overall image;
- FIG. 2 is a diagram for explaining a
- FIG. 4 is a diagram for explaining a method of calculating a contribution based on the statistical model of FIG. 3;
- FIG. 6 is a flow chart for explaining existence position detection processing executed by a control unit in the first embodiment.
- FIG. 11 is a first flow chart for explaining a position detection process executed by a control unit in the second embodiment;
- FIG. 11 is a first flow chart for explaining existence position detection processing executed by a control unit in the third embodiment;
- FIG. 1 shows a configuration example of an object operation system 11 including an image processing device 10 according to the first embodiment.
- the object manipulation system 11 may comprise a camera 12 , an image processing device 10 and a manipulation device 13 .
- the target operation system 11 performs a specific operation on a detection target 14 such as a product in a factory or the like, for example.
- the detection target 14 is placed, for example, on a tray 15 or the like.
- the camera 12 may image the tray 15 on which the detection target 14 is placed.
- the camera 12 may, for example, generate images at a predetermined frame rate, such as 30 fps, in other words periodically.
- the image processing device 10 may detect the position of the detection target 14 using the image captured by the camera 12 .
- the operation device 13 may operate the detection target 14 at the position detected by the image processing device 10 .
- the operation device 13 is, for example, a robot, and may grip the detection target 14 as an operation.
- the image processing apparatus 10 includes a communication unit (acquisition unit) 16, a memory 17, and a control unit 18.
- the image processing device 10 may further include an input section 19 .
- the communication unit 16 acquires the entire image from the camera 12.
- the whole image is an image from which the position of the detection target 14 is detected by the image processing device 10 .
- the whole image is an image including a subject to be matched.
- the target to be matched is a target to be matched with the detection target of the representative image, as will be described later.
- the communication unit 16 may be, for example, a communication interface that communicates information and commands by communicating with an external device.
- the communication unit 16 may provide the operation device 13 with the existing position of the detection target 14 in the entire image.
- the memory 17 includes arbitrary storage devices such as RAM (Random Access Memory) and ROM (Read Only Memory).
- the memory 17 may store various programs that cause the control unit 18 to function and various information that the control unit 20 uses.
- the memory 17 stores the template image group.
- a template image group includes a plurality of detection target images.
- the plurality of detection target images may be acquired in chronological order as described later.
- the plurality of detection target images may include, for example, a plurality of detection target images generated by padding processing from one image.
- the plurality of detection target images may include, for example, a plurality of detection target images acquired under different environments.
- a template image group including a plurality of detection target images acquired in chronological order will be described as an example.
- the detection target image is an image of the detection target 14 .
- the plurality of detection target images may include a plurality of processed images obtained by performing image processing on some of the plurality of detection target images.
- Image processing may be performed by the control unit 18 .
- the image processing is performed by, for each of a plurality of detection target images, detecting changes in pixel values between a plurality of detection target images obtained before and after the detection target image in the template image group. This is luminance adjustment processing applied to the entire image.
- the image processing is, for example, a process of adding a disturbance suddenly occurring to a plurality of detection target images before and after each of a part of the plurality of detection target images.
- the input unit 19 may include one or more interfaces that detect user's operation input.
- the input unit 19 may include, for example, physical keys, capacitive keys, and a touch screen provided integrally with a display device.
- the control unit 18 includes one or more processors and memory.
- the processor may include a general-purpose processor that loads a specific program to perform a specific function, and a dedicated processor that specializes in specific processing.
- a dedicated processor may include an Application Specific Integrated Circuit (ASIC).
- the processor may include a programmable logic device (PLD).
- the PLD may include an FPGA (Field-Programmable Gate Array).
- the control unit 18 may be either an SoC (System-on-a-Chip) in which one or more processors cooperate, or a SiP (System In a Package).
- the control unit 18 generates a representative image based on a partial image group extracted from the template image group. Extraction of a partial image group from a template image group and generation of a representative image will be explained later using a detailed example. For example, the control unit 18 may generate a representative image each time an entire image is acquired.
- the control unit 18 superimposes a representative image ri having a size equal to or smaller than the overall image wi on at least a partial area of the overall image wi, and determines the similarity between the representative image ri and the area. I do.
- the control unit 18 can detect the existence position of the detection target, for example, along with determining whether or not the matching target is the detection target.
- the control unit 18 stores the image in the memory 17 so as to be included in the template image group as the detection target image.
- the control unit 18 may determine, for example, the first time point that is the start time point of the temporal range regarding the acquisition time of the detection target image extracted from the template image group.
- the first time point may be determined in any manner.
- the first point in time may be determined, for example, based on changes in appearance effects occurring on images included in the template image group over time.
- the input unit 19 detects an operation input designating the first time point, the time point corresponding to the operation input may be determined as the first time point.
- the first time point may be determined based on analysis of the template images by the controller 18 .
- the control unit 18 determines the first point in time based on, for example, time-series changes in the pixel values of each image included in the template image group.
- the pixel value of each image may be, for example, a pixel value of a specific pixel forming the image, an average value of pixel values of all images, an average value of pixel values in a specific range, or the like. More specifically, the control unit 18 calculates, for example, a period that is at least twice the period corresponding to the lowest frequency component in changes in pixel values along the time series, and causes the calculated period to go back from the current point. is determined to be the first time point.
- the control unit 18 designates an image group including a plurality of images acquired from a first time in the past to the latest time in chronological order as a first image. You can define it as a group.
- the control unit 18 may determine an image sampling interval for extracting further images from the first image group.
- a sampling interval may be determined in any manner. The sampling interval may be determined, for example, based on the time-series change in appearance effects on the images included in the template image group.
- the input unit 19 detects an operation input specifying a sampling interval, the interval corresponding to the operation input may be determined as the sampling interval. Alternatively, the sampling interval may be determined based on analysis of the template image group by the control unit 18 .
- the control unit 18 determines the sampling interval, for example, based on the time-series change in the pixel value of each image included in the template image group.
- the pixel value of each image is the pixel value of a specific pixel constituting the image, the average pixel value of all images, the average pixel value of a specific range, etc. good. More specifically, the control unit 18 determines, for example, a sampling interval that is twice as long as the period corresponding to the highest frequency component among changes in pixel values along the time series.
- the control unit 18 may determine the sampling interval to be any interval equal to or longer than the acquisition cycle of the detection target image. Note that the fact that the absolute value of the difference between the pixel values of consecutive time series is less than the first threshold may be regarded as the fact that the pixel values of the images included in the template image group do not change in time series. Further, for example, when the pixel values of the images included in the template image group do not change in time series, the control unit 18 sets the number of detection target images detected from the first image group at the above sampling interval to a predetermined number. The first time point may be determined to exceed the sampling number of .
- control unit 18 may define an image group including a plurality of images acquired at sampling intervals from the first time point as the second image group.
- the control unit 18 may extract further images from the second group of images, as described below.
- the control unit 18 detects at least one disturbance factor with respect to the detection target image included in the template image group by detecting the existing position from the entire image wi within a predetermined time range going back from the present time. You can decide.
- the detection target images included in the template image group whose existing positions are detected within a predetermined time range going back from the present may be detection target images included in each of the entire images wi of a plurality of frames.
- the disturbance factors in the plurality of detection target images respectively corresponding to the plurality of whole images wi may be the same or similar.
- a disturbance factor is a factor that affects the appearance.
- Disturbance factors include, for example, changes over time in the direction and intensity of illumination due to sunlight, changes in received colors due to changes in the position of the camera 12, presence or absence of labeling on the detection target 14 or differences in position, illumination and detection target. 14, illuminance changes due to the passage of workers or robots, and the like.
- the disturbance factor may be determined by any method. For example, when the input unit 19 detects an operation input that designates a situation that causes an effect on the appearance in the newly acquired entire image wi, the control unit 18 sets the factor corresponding to the situation to the disturbance factor. decide.
- the control unit 18 may determine a plurality of disturbance factors when detecting an operation input designating a plurality of situations. Alternatively, a single or multiple disturbance factors may be determined based on analysis of the second group of images by the control unit 18 .
- control unit 18 may determine the disturbance factor by generating a comparison image in which the disturbance factor is reflected.
- the control unit 18 may generate a comparison image using detection target images within a predetermined time range going back from the present. For example, when the detection target image is single, the comparison image is the detection target image itself. For example, when there are a plurality of detection target images, the comparison image may be generated based on the plurality of detection target images. A comparison image based on a plurality of detection target images may be generated, for example, by averaging the plurality of detection target images.
- the control unit 18 may extract an image group of detection target images corresponding to the determined disturbance factor from the second image group. When a plurality of disturbance factors are determined, the control unit 18 may extract a plurality of image groups of detection target images corresponding to each disturbance factor.
- the control unit 18 may extract, from the second image group, an image group of detection target images associated with a situation corresponding to the disturbance factor.
- the control unit 18 compares each detection target image included in the second image group with the comparison image to determine the detection target image corresponding to the comparison image. image group.
- control unit 18 compares the pixel values of each pixel forming the comparison image and each pixel forming each detection target image included in the second image group, and detects images that are determined to be similar to the comparison image. An image group of target images is extracted. Further, for example, the control unit 18 clusters the image groups included in the comparative image and the second image group, and extracts the image group of the detection target images belonging to the same group as the comparative image. Further, for example, the control unit 18 performs disturbance determination on all detection target images included in the second image group, and extracts an image group of the detection target images including pixels included in the comparison image and determined to be disturbance. .
- the control unit 18 may generate the representative image ri based on the second image group extracted from the template image group or based on the image group corresponding to the disturbance factor further extracted from the second image group.
- the control unit 18 may generate the representative image ri by various methods.
- control unit 18 generates the representative image ri by setting a statistic based on the pixel values of the pixels at the addresses corresponding to each other among the images included in the second image group to the pixel values of the pixels.
- a statistic is, for example, an average value, a mode value, a median value, or the like.
- pixels at addresses corresponding to each other may be only pixels located at the same address between images.
- the pixels at addresses corresponding to each other may be, as shown in FIG. 4, a plurality of pixels included in, for example, a 3 ⁇ 3 pixel area centered on the same address.
- the pixels at addresses corresponding to each other may be a plurality of pixels in a section including the same address among the sections obtained by dividing each image into a plurality of grids, as shown in FIG.
- the statistics may include a plurality of detection target images including a plurality of processed images described above.
- control unit 18 may generate the representative image ri by performing disturbance determination on each pixel in each image included in the image group corresponding to a single disturbance factor.
- the single disturbance factor affects the pixel values in all the images for which disturbance determination is performed here. Therefore, the single disturbance factor is not regarded as a disturbance in the image for which disturbance determination is performed.
- the disturbance determination performed here is a determination as to whether or not a sudden disturbance has occurred in each image included in the image group corresponding to the determined disturbance factor.
- the control unit 18 creates a histogram for pixel values of pixels at mutually corresponding addresses in each image, and determines pixel values whose frequency is equal to or higher than a threshold value to be non-disturbance.
- the control unit 18 approximates the pixel values of pixels at mutually corresponding addresses in each image to a Gaussian distribution, and determines pixel values whose probability density is equal to or higher than a threshold value to be non-disturbance.
- the control unit 18 determines the disturbance by classifying the pixel values of the pixels at the mutually corresponding addresses in each image into clusters using a statistical model such as K-Means or mixed Gaussian distribution.
- the control unit 18 generates the representative image ri by setting the statistic based on the pixel value determined as non-disturbing to the pixel value of the pixel.
- a statistic is, for example, an average value, a mode value, a median value, or the like.
- control unit 18 generates a representative image ri by selecting a single detection target image based on the disturbance factor from among the image group corresponding to the single disturbance factor. For example, the control unit 18 selects, as the representative image ri, the detection target image having the smallest pixel value difference from the statistic based on the pixel values of pixels at mutually corresponding addresses.
- the control unit 18 determines the largest disturbance factor in the comparative image among the plurality of disturbance factors.
- the maximum disturbance factor may be defined arbitrarily.
- the maximum disturbance factor is, for example, the disturbance factor with the largest number of images included in the corresponding image group, the disturbance factor with the largest number of corresponding pixels, or the disturbance factor with the latest occurrence time.
- the control unit 18 calculates a statistic based on pixel values determined to be non-disturbance, or selects a single detection target image in the image group. Thus, a representative image ri is generated.
- the control unit 18 when extracting a plurality of image groups corresponding to a plurality of disturbance factors, the control unit 18 generates a plurality of primary representative images for each of the plurality of image groups.
- a primary representative image may be generated in the same manner as the above-described representative image ri generated based on a group of images corresponding to a single disturbance factor.
- the control unit 18 may generate the representative image ri by replacing the plurality of detection target images with the plurality of primary representative images in the method of generating the representative image ri based on the plurality of detection target images described above.
- the control unit 18 may generate, for example, a statistic based on pixel values of pixels at the same address in each of the plurality of primary representative images by determining the pixel values.
- the control unit 18 calculates a plurality of statistics based on pixel values of pixels at mutually corresponding addresses in each of a plurality of image groups corresponding to each of a plurality of disturbance factors.
- the control unit 18 generates the representative image ri by determining the pixel value of each pixel based on a plurality of statistics. For example, when a plurality of statistic amounts corresponding to each of a plurality of image groups are partially common to pixels at mutually corresponding addresses, the control unit 18 may determine the common statistic amount as the pixel value of the pixel. That some of the plurality of statistics are common may mean not only that the statistics match, but also that the statistics fall within a predetermined range.
- a statistic is, for example, an average value, a mode value, a median value, or the like.
- control unit 18 performs similarity determination by comparing the generated representative image ri with a partial area of the entire image wi.
- the control unit 18 compares the pixel value of at least a part of all the pixels forming the generated representative image ri with the pixel value of the pixel at the same address as that pixel in a part of the whole image wi. Similar judgment may be made by
- the control unit 18 may compare all pixels forming the generated representative image ri with all pixels in a partial area of the entire image wi.
- control unit 18 may select pixels to be used for comparison with the entire image wi from among a plurality of pixels forming the generated representative image ri.
- control unit 18 may determine feature points such as edges and corners as pixels used for comparison in the representative image ri.
- control unit 18 may perform similarity determination using the degree of contribution calculated for each of a plurality of pixels forming the representative image ri or for each of a plurality of pixels forming a partial region of the entire image wi.
- the degree of contribution is an index that indicates the degree to which the corresponding pixel contributes to similarity determination.
- the control unit 18 may calculate the degree of contribution based on statistics of at least part of the template image group. At least part of the template image used for calculating the contribution may be the image group used for generating the representative image ri. Alternatively, at least part of the template image used for calculating the degree of contribution is a group of images corresponding to disturbance factors other than the selected single disturbance factor among the plurality of disturbance factors determined for the entire image wi. may be at least part of
- the control unit 18 calculates the pixel value of at least one pixel corresponding to each address in a plurality of detection target images of at least a part of the template image group. Create statistical models such as histograms or Gaussians. As shown in FIG. 3, the at least one pixel corresponding to the address may be the only pixel located at the address as shown in FIG. Alternatively, as shown in FIG. 4, at least one pixel corresponding to the address may be a plurality of pixels included in, for example, a 3 ⁇ 3 pixel area centered on the address. Alternatively, as shown in FIG.
- At least one pixel corresponding to the address may be a plurality of pixels in a section including the address among sections obtained by dividing the detection target image into a plurality of grids.
- the statistical model may include a plurality of detection target images including a plurality of processed images described above.
- the control unit 18 calculates the frequency corresponding to the pixel value of the pixel at each address in the partial area of the entire image wi corresponding to the pixel at each address used for the comparison determined as described above, in the statistical model of the pixel. Read out. As shown in FIG. 6, the control unit 18 calculates the contribution of the pixel according to the frequency. The contribution may be calculated so as to increase with frequency.
- the control unit 18 selects pixels used for similarity determination from among a plurality of pixels forming the representative image ri or a plurality of pixels forming a part of the entire image wi, corresponding to the pixels used for comparison. You can For example, the control unit 18 uses pixels whose contribution degree is equal to or greater than a threshold value for similarity determination. Alternatively, the control unit 18 may not perform similarity determination based on similarity for pixels whose contribution is equal to or less than the threshold. Specifically, for example, if a pixel used for comparison of representative images ri has a pixel value with low frequency in the corresponding statistical model, or has a pixel value outside a certain range centered on the pixel value with the highest frequency.
- control unit 18 may perform weighting according to the degree of contribution in similarity determination. For example, the control unit 18 may multiply the difference between the pixels at the same address in the representative image ri and the partial area of the entire image wi by a weight that changes according to the degree of contribution.
- the control unit 18 may calculate the degree of similarity by summing the differences between pixels forming the generated representative image ri and pixels in a partial region of the entire image wi, which have the same address.
- the control unit 18 may determine a partial area of the entire image wi whose similarity is equal to or less than the threshold as the presence position of the detection target 14 .
- control unit 18 may not perform similarity judgment based on the degree of similarity when a specific condition is satisfied using the created statistical model of pixels for calculating the degree of contribution.
- a partial area of the entire image wi may be determined to be dissimilar to the representative image.
- a specific condition is when, among all the pixels used for comparison in the partial region, the number of pixels with low frequency pixel values in the corresponding statistical model is large, or when the number of pixels with the most frequency pixel values is constant This is the case where there are a large number of pixels having pixel values outside the range of .
- the pixel value for example, in the statistical model corresponding to each pixel, if the pixel value frequency of the pixel is equal to or less than a threshold, the pixel value may be determined to be a low-frequency pixel value. Further, for example, in a statistical model approximated by a Gaussian model, when the frequency of the pixel value of the pixel is located outside a certain range centered on the pixel value of the mode, the pixel value is It may be determined that the pixel value is out of a certain range centered on the most frequent pixel value. When the statistical model is approximated to a Gaussian mixture model, the certain range may be set to a range of ⁇ 6 ⁇ around the mode of each Gaussian distribution, for example.
- a large number of pixels with low frequency pixel values among all pixels may mean that the number of pixels with low frequency pixel values is equal to or less than a threshold.
- a large number of pixels with low frequency pixel values among all pixels may mean that the ratio of pixels with low frequency pixel values to all pixels used for comparison is equal to or less than a threshold.
- the control unit 18 sets the image portion of the partial area of the entire image wi corresponding to the existence position as the detection target image to the template image group. , and stored in the memory 17.
- the control unit 18 determines the disturbance factor based on the operation input detected by the input unit 19 and detects the presence position of the detection target 14 in the entire image wi, the detection target image corresponding to the presence position has an external appearance Influencing situations may be associated and included in the set of template images.
- the existence position detection processing executed by the control unit 18 in the first embodiment will be described using the flowchart of FIG.
- the existence position detection processing is started by acquiring the entire image wi of one frame.
- step S100 the control unit 18 temporarily identifies a partial area having the same size as the detection target image included in the template image group in the acquired overall image wi as an area for similarity determination. After identification, the process proceeds to step S101.
- step S101 the control unit 18 determines the first time point and the sampling period based on the detection by the input unit 19 or the analysis of each detection target image included in the template image group. After determination, the process proceeds to step S102.
- step S102 the control unit 18 extracts a second image group based on the first time point and sampling interval determined in step S101. After extraction, the process proceeds to step S103.
- step S103 the control unit 18 determines the disturbance factor based on the input designating the situation to the input unit 19 or the analysis of the second image group. After determination, the process proceeds to step S104.
- step S104 the control unit 18 generates a representative image ri based on the image group corresponding to the disturbance factor determined in step S103, extracted from the second image group extracted in step S102. After generating the representative image ri, the process proceeds to step S105.
- step S105 the control unit 18 selects pixels to be used for comparison with the partial area identified in step S100 in the representative image ri generated in step S104. After selecting a pixel, the process proceeds to step S106.
- step S106 the control unit 18 extracts part of the template image group. After extraction, the process proceeds to step S107.
- control unit 18 creates a statistical model for each pixel of the plurality of detection target images extracted at step S106. After creating the statistical model, the process proceeds to step S108.
- step S108 the control unit 18 uses the statistical model created in step S107 to determine whether the number of pixels with low frequency pixel values is large. If the number of pixels with low frequency is not large, the process proceeds to step S109. If the number of pixels with low frequency is large, the process proceeds to step S114.
- step S109 the control unit 18 calculates the degree of contribution by reading the frequencies corresponding to the pixel values of the pixels forming the partial region identified in step S100 from the statistical model created in step S107. After calculating the contribution, the process proceeds to step S110.
- step S110 the control unit 18 calculates the similarity of the partial region identified in step S100 to the representative image ri based on the contribution calculated in step S109. After calculating the similarity, the process proceeds to step S111.
- step S111 the control unit 18 determines whether or not the similarity calculated in step S110 is equal to or less than the threshold. If so, the process proceeds to step S112. Otherwise, the process proceeds to step S114.
- step S112 the control unit 18 determines that the partial area identified in step S100 is the position where the detection target 14 exists. After determination, the process proceeds to step S113.
- step S113 the control unit 18 stores the image of the partial area identified in step S100 in the memory 17 as a detection target image so as to be included in the template image group. After storing, the process proceeds to step S114.
- step S114 the control unit 18 determines whether or not the entire area of the entire image wi has been specified as a partial area. If all regions have not been identified, the process proceeds to step S115. If all areas have been specified, the detection process ends.
- step S115 the control unit 18 displaces the partial area identified in step S100 while maintaining the same size. After displacement, the process returns to step S100.
- the image processing apparatus 10 of the first embodiment acquires a representative image ri based on a part of the template image group, and determines similarity between at least part of the entire image wi and the representative image ri. conduct.
- the image processing apparatus 10 can generate a representative image ri including a detection target image including disturbances that occur frequently. Therefore, the image processing apparatus 10 can include the influence of a disturbance with a high frequency of occurrence in the representative image ri even if it is a disturbance as a whole. It can be determined whether or not the subject corresponds to the detection target 14 .
- the image processing apparatus 10 can continuously and robustly detect the detection target 14 using a relatively small number of learning images, in other words, a group of images, even if the frequency of occurrence of disturbances is high.
- the image processing apparatus 10 of the first embodiment stores the detection target image of the detection target 14 detected within the entire image wi in the memory 17 so as to be included in the template image group.
- the image processing apparatus 10 can reflect, in the representative image ri, the tendency of disturbances that are likely to affect the entire image wi in subsequent detection of the existence position of the detection target 14 . Therefore, the image processing apparatus 10 can improve robustness of detection in subsequent detection of the existence position of the detection target 14 .
- the image processing apparatus 10 of the first embodiment includes a first image including a plurality of images acquired in chronological order from a first point in time going back in time, among the detection target images included in the template image group.
- a representative image ri is generated based on a second image group including a plurality of images acquired at sampling intervals.
- the image processing apparatus 10 uses all the detection target images included in the template image group by determining the sampling interval according to the shortest generation cycle of the disturbances that affect when the entire image wi is captured.
- An appropriate representative image ri can be generated using a portion of the detection target image without the need. Therefore, the image processing apparatus 10 can reduce the processing load for generating the representative image ri, and thus can reduce the load associated with detecting the existing position of the detection target 14 .
- the image processing apparatus 10 of the first embodiment determines at least one disturbance factor that has the same or similar appearance effect on the detection target 14 included in the plurality of whole images wi, An image group corresponding to the disturbance factor is extracted from the group, and a representative image ri is generated based on the image group corresponding to the disturbance factor.
- the image processing apparatus 10 can reflect in the representative image ri the disturbance that is highly likely to affect the acquired representative image ri. Therefore, the image processing device 10 can further improve the robustness of detection of the detection target 14 .
- the situation in response to an operation input of a situation that affects the appearance of the detection target 14, the situation is associated with the situation in the template image group.
- a disturbance factor is determined based on the detection target image, and when detecting the presence position of the detection target 14 in the entire image wi, the situation is associated with the presence position image and included in the template image group.
- the image processing apparatus 10 of the first embodiment selects pixels to be used for comparison with the overall image wi from among the plurality of pixels forming the representative image ri. With such a configuration, the image processing apparatus 10 can reduce the load of similarity determination processing compared to a configuration in which all pixels forming the representative image ri are compared with a partial region of the entire image wi.
- the image processing apparatus 10 of the first embodiment generates a representative image ri by setting a statistic based on the pixel values of pixels at the same address in each image group corresponding to the disturbance factor to the pixel values of the pixels.
- the image processing apparatus 10 can remove the influence of pixel values that can be regarded as disturbances in the image group corresponding to the disturbance factor, in other words, pixel values that occur infrequently in the image group. . Therefore, the image processing device 10 can further improve the robustness of detection of the detection target 14 .
- the image processing apparatus 10 calculates the degree of contribution of a part of the whole image wi used for similarity determination based on the statistics of at least part of the detection target image of the template image group, and calculates the contribution degree is used for similarity judgment.
- the image processing apparatus 10 can reduce the possibility that the similarity determination results of pixels with a low degree of contribution will be determined to have high similarity. Therefore, when a disturbance that does not occur in the template image group occurs in the entire image wi, the image processing apparatus 10 can eliminate the influence of the disturbance in similarity determination. Therefore, the image processing device 10 can further improve the robustness of detection of the detection target 14 .
- the image processing apparatus 10 of the first embodiment at least part of the template image group used for calculating the contribution is the image group used for generating the representative image ri.
- the image processing apparatus 10 can eliminate the influence of the disturbance in the similarity determination when a disturbance that is not reflected in the representative image ri occurs in the entire image wi. Therefore, the image processing device 10 can further improve the robustness of detection of the detection target 14 .
- the image processing apparatus 10 of the first embodiment at least a part of the template image group used for calculating the degree of contribution is selected from among the plurality of disturbance factors determined for the entire image wi. It is at least part of an image group corresponding to a disturbance factor other than one disturbance factor.
- the image processing apparatus 10 when a disturbance factor that may not be reflected in the representative image ri and has occurred occurs in the entire image wi, is reflected in the representative image ri in similarity determination. It is possible to eliminate the influence of the disturbance factor that is not Therefore, the image processing device 10 can further improve the robustness of detection of the detection target 14 .
- the representative image ri is generated based on the image group corresponding to the disturbance factor extracted from the second image group, but the configuration is not limited to this.
- the representative image ri may be generated from the second image group without specifying the image group corresponding to the disturbance factor.
- the second embodiment differs from the first embodiment in the method of extracting an image group from a template image group in the control unit 18 .
- the second embodiment will be described below, focusing on the differences from the first embodiment.
- symbol is attached
- an image processing apparatus 10 includes a communication section 16, a memory 17, and a control section 18, as in the first embodiment. Further, the image processing apparatus 10 may include an input unit 19, as in the first embodiment.
- the configurations and functions of the communication unit 16, memory 17, and input unit 19 in the second embodiment are the same as in the first embodiment.
- the configuration of the control unit 18 in the second embodiment is the same as in the first embodiment.
- a detection target image whose position is detected within a predetermined time range going back from the present and included in the template image group may be a detection target image in at least one frame of the entire image wi.
- a plurality of detection target images in each of the entire images wi of a plurality of frames. Disturbance factors in a plurality of detection target images may be the same or similar.
- control unit 18 may generate the representative image ri based on the image group corresponding to the disturbance factor extracted from the template image group.
- a method of generating the representative image ri based on the image group corresponding to the disturbance factor may be similar to that of the first embodiment.
- the control unit 18 performs similarity determination by comparing the generated representative image ri with a partial area of the entire image wi. Therefore, the control unit 18 may compare all pixels forming the generated representative image ri with all pixels in a partial region of the entire image wi. Alternatively, the control unit 18 may select pixels to be used for comparison with the entire image wi from among a plurality of pixels forming the generated representative image ri. Furthermore, the control unit 18 may perform similarity determination using the degree of contribution calculated for each pixel that constitutes a partial area of the entire image wi.
- the existence position detection processing executed by the control unit 18 in the second embodiment will be described using the flowchart of FIG.
- the existence position detection processing is started by acquiring the entire image wi of one frame.
- step S200 the control unit 18 receives an input designating a situation to the input unit 19, or a position detected from the entire image wi within a predetermined time range going back from the present and included in the template image group.
- a disturbance factor is determined based on the analysis of the target image. After determination, the process proceeds to step S201.
- step S201 the control unit 18 extracts an image group corresponding to the disturbance factor determined in step S200 from the template image group. After extraction, the process proceeds to step S202.
- control unit 18 performs processing similar to steps S104 to S115 in the detection processing of the first embodiment.
- the image processing apparatus 10 of the second embodiment configured as described above also generates a representative image ri based on a partial group of images extracted from a group of template images, and combines a part of the entire image wi with the representative image ri. , the presence position of the detection target 14 in the entire image wi is detected. Therefore, the image processing apparatus 10 can continuously and robustly detect the detection target 14 using a relatively small number of learning images, in other words, a group of images, even if the frequency of occurrence of disturbances is high.
- the third embodiment differs from the first embodiment in the method of extracting the image group from the template image group in the control unit 18 .
- the third embodiment will be described below, focusing on the differences from the first embodiment.
- symbol is attached
- an image processing apparatus 10 includes a communication section 16, a memory 17, and a control section 18, as in the first embodiment. Further, the image processing apparatus 10 may include an input unit 19, as in the first embodiment.
- the configurations and functions of the communication unit 16, memory 17, and input unit 19 in the third embodiment are the same as in the first embodiment.
- the configuration of the control unit 18 in the third embodiment is the same as in the first embodiment.
- control unit 18 extracts arbitrary image groups from the template image group, not limited to the second image group and the image group corresponding to the disturbance factor.
- the control unit 18 may generate the representative image ri based on the extracted image group.
- a method of generating the representative image ri based on the extracted image group may be similar to the method of generating the representative image ri based on the second image group in the first embodiment.
- the control unit 18 performs similarity determination by comparing the generated representative image ri with a partial area of the entire image wi. Therefore, the control unit 18 may compare all pixels forming the generated representative image ri with all pixels in a partial region of the entire image wi. Alternatively, the control unit 18 may select pixels to be used for comparison with the entire image wi from among a plurality of pixels forming the generated representative image ri. Note that in the third embodiment, the control unit 18 calculates the degree of contribution calculated for each pixel constituting a partial region of the entire image wi based on the statistics of at least a portion of the template image group, Similarity determination is performed using the contribution.
- the existing position detection processing is started by obtaining a whole image wi of one frame.
- step S300 the control unit 18 extracts an arbitrarily determined image group from the template image group. After extraction, the process proceeds to step S301.
- control unit 18 performs processing similar to steps S104 to S115 in the detection processing of the first embodiment.
- the image processing apparatus 10 of the third embodiment configured as described above generates a representative image ri based on a partial image group extracted from the template image group, and a part of the entire image wi and the representative image ri When detecting the existence position of the detection target 14 in the whole image wi by similarity judgment with , the contribution is used for similarity judgment.
- the image processing apparatus 10 can eliminate the influence of the disturbance in the similarity judgment when a disturbance that is not reflected in the representative image ri occurs in the entire image wi. Therefore, the image processing device 10 can further improve the robustness of detection of the detection target 14 .
- a storage medium in which the program is recorded for example, an optical disk, a magneto-optical A disk, CD-ROM, CD-R, CD-RW, magnetic tape, hard disk, memory card, etc.
- an optical disk, a magneto-optical A disk, CD-ROM, CD-R, CD-RW, magnetic tape, hard disk, memory card, etc. can also be used.
- the implementation form of the program is not limited to an application program such as an object code compiled by a compiler or a program code executed by an interpreter. good.
- the program may or may not be configured so that all processing is performed only in the CPU on the control board.
- the program may be configured to be partially or wholly executed by another processing unit mounted on an expansion board or expansion unit added to the board as required.
- Embodiments according to the present disclosure are not limited to any specific configuration of the embodiments described above. Embodiments of the present disclosure extend to any novel feature or combination thereof described in the present disclosure or any novel method or process step or combination thereof described. be able to.
- Descriptions such as “first” and “second” in this disclosure are identifiers for distinguishing the configurations. Configurations that are differentiated in descriptions such as “first” and “second” in this disclosure may interchange the numbers in that configuration. For example, a first point in time can exchange identifiers “first” and “second” with a second point in time. The exchange of identifiers is done simultaneously. The configurations are still distinct after the exchange of identifiers. Identifiers may be deleted. Configurations from which identifiers have been deleted are distinguished by codes. The description of identifiers such as “first” and “second” in this disclosure should not be used as a basis for interpreting the order of the configuration or the existence of lower numbered identifiers.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
Description
検出対象の複数の画像を含むテンプレート画像群を記憶するメモリと、
照合対象を含む全体画像を取得する取得部と、
前記テンプレート画像群の一部に基づいた代表画像を取得し、前記全体画像の少なくとも一部と前記代表画像との類似判断を行なう制御部と、を備える。
検出対象の複数の画像を含むテンプレート画像群を記憶するメモリと、
照合対象を含む全体画像を取得する取得部と、
前記テンプレート画像群の一部に基づいた代表画像を取得し、前記全体画像の少なくとも一部と前記代表画像との類似判断を行なう制御部と、を備え、
前記制御部は、前記類似判断に用いる前記全体画像の一部の寄与度を前記テンプレート画像群の少なくとも一部の統計量に基づいて算出し、該寄与度を前記類似判断に用いる
検出対象の複数の画像を含むテンプレート画像群を記憶することと、
照合対象を含む全体画像を取得することと、
前記テンプレート画像群の一部に基づいた代表画像を取得することと、
前記全体画像の少なくとも一部と前記代表画像との類似判断を行うことと、を備える。
11 対象操作システム
12 カメラ
13 操作機器
14 検出対象
15 トレイ
16 通信部
17 メモリ
18 制御部
19 入力部
ri 代表画像
wi 全体画像
Claims (19)
- 検出対象の複数の画像を含むテンプレート画像群を記憶するメモリと、
照合対象を含む全体画像を取得する取得部と、
前記テンプレート画像群の一部に基づいた代表画像を取得し、前記全体画像の少なくとも一部と前記代表画像との類似判断を行なう制御部と、を備える
画像処理装置。 - 請求項1に記載の画像処理装置において、
前記制御部は、前記全体画像内で検出された前記検出対象の画像を前記テンプレート画像群に含めるように前記メモリに格納する
画像処理装置。 - 請求項1又は2に記載の画像処理装置において、
前記制御部は、前記テンプレート画像群に含まれる画像のうち、過去に遡った第1の時点から時系列に沿って取得された複数の画像を含む第1画像群のうち、サンプリング間隔で取得された複数の画像を含む第2画像群に基づいて前記代表画像を生成する
画像処理装置。 - 請求項3に記載の画像処理装置において、
前記第1の時点及び前記サンプリング間隔の少なくとも一方の入力を検出する入力部を、更に備え、
前記制御部は、前記入力部が検出する入力に基づいて、前記第1の時点及び前記サンプリング間隔の少なくとも一方を決定する
画像処理装置。 - 請求項1に記載の画像処理装置において、
前記制御部は、複数の前記全体画像に含まれる前記検出対象に対して同一又は類似する外観上の影響を与える少なくとも一つの外乱要因を決定し、前記テンプレート画像群から該外乱要因に対応する画像群を抽出し、該外乱要因に対応する画像群に基づいて前記代表画像を生成する
画像処理装置。 - 請求項3又は4に記載の画像処理装置において、
前記制御部は、複数の前記全体画像に含まれる前記検出対象に対して同一又は類似する外観上の影響を与える少なくとも一つの外乱要因を決定し、前記第2画像群から該外乱要因に対応する画像群を抽出し、該外乱要因に対応する画像群に基づいて前記代表画像を生成する
画像処理装置。 - 請求項5又は6に記載の画像処理装置において、
前記全体画像に含まれる前記検出対象に対する前記影響が生じる時の状況の入力を検出する入力部を、更に備え、
前記制御部は、前記外乱要因を前記テンプレート画像群の中で前記状況が関連付けられている画像に基づいて決定し、前記全体画像における前記検出対象の存在位置を検出する場合に前記存在位置の画像に前記状況を関連付けてテンプレート画像群に含める
画像処理装置。 - 請求項1から7のいずれか1項に記載の画像処理装置において、
前記制御部は、前記代表画像を構成する複数の画素の中から、前記全体画像との比較に用いる画素を選択する
画像処理装置。 - 請求項5から7のいずれか1項に記載の画像処理装置において、
前記制御部は、前記外乱要因に対応する画像群それぞれの同一アドレスの画素の画素値に基づく統計量を該画素の画素値に定めることにより前記代表画像を生成する
画像処理装置。 - 請求項5から7のいずれか1項に記載の画像処理装置において、
前記制御部は、前記外乱要因に対応する画像群の中で、前記外乱要因に基づいて単一の画像を選択することにより、前記代表画像を生成する
画像処理装置。 - 請求項5から8のいずれか1項に記載の画像処理装置において、
前記制御部は、前記全体画像に対する複数の外乱要因を決定し、前記複数の外乱要因それぞれに対応する複数の画像群を抽出し、前記複数の画像群別に複数の一次代表画像を生成し、該複数の一次代表画像それぞれの同一アドレスの画素の画素値に基づく統計量を該画素の画素値に定めることにより前記代表画像を生成する
画像処理装置。 - 請求項5から8のいずれか1項に記載の画像処理装置において、
前記制御部は、前記全体画像に対する複数の外乱要因を決定し、前記複数の外乱要因それぞれに対応する複数の画像群を抽出し、前記複数の画像群それぞれにおいて同じアドレスの画素の画素値に基づく複数の統計量を算出し、前記複数の統計量に基づいて該画素の画素値を定めることにより前記代表画像を生成する
画像処理装置。 - 請求項5から12のいずれか1項に記載の画像処理装置において、
前記制御部は、前記類似判断に用いる前記全体画像の一部の寄与度を前記テンプレート画像群の少なくとも一部の統計量に基づいて算出し、該寄与度を前記類似判断に用いる
画像処理装置。 - 請求項13に記載の画像処理装置において、
前記制御部は、前記全体画像の一部を構成する画素別に前記寄与度を算出し、前記寄与度に基づいて該一部を構成する画素の中から前記類似判断に用いる画素を選択する
画像処理装置。 - 請求項13に記載の画像処理装置において、
前記制御部は、前記全体画像の一部を構成する画素別に前記寄与度を算出し、前記類似判断において前記画素別に算出された寄与度に応じた重み付けを施す
画像処理装置。 - 請求項13から15のいずれか1項に記載の画像処理装置において、
前記寄与度の算出に用いる前記テンプレート画像群の少なくとも一部は、前記代表画像の生成に用いた画像群である
画像処理装置。 - 請求項13から15のいずれか1項に記載の画像処理装置において、
前記寄与度の算出に用いる前記テンプレート画像群の少なくとも一部は、前記全体画像に対して決定された複数の外乱要因の中で、選択された単一の外乱要因以外の外乱要因に対応する画像群の中の少なくとも一部である
画像処理装置。 - 検出対象の複数の画像を含むテンプレート画像群を記憶するメモリと、
照合対象を含む全体画像を取得する取得部と、
前記テンプレート画像群の一部に基づいた代表画像を取得し、前記全体画像の少なくとも一部と前記代表画像との類似判断を行なう制御部と、を備え、
前記制御部は、前記類似判断に用いる前記全体画像の一部の寄与度を前記テンプレート画像群の少なくとも一部の統計量に基づいて算出し、該寄与度を前記類似判断に用いる
画像処理装置。 - 検出対象の複数の画像を含むテンプレート画像群を記憶することと、
照合対象を含む全体画像を取得することと、
前記テンプレート画像群の一部に基づいた代表画像を取得することと、
前記全体画像の少なくとも一部と前記代表画像との類似判断を行うことと、を備える
画像処理方法。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202280058115.9A CN117897739A (zh) | 2021-08-27 | 2022-07-08 | 图像处理设备和图像处理方法 |
EP22860997.0A EP4394697A1 (en) | 2021-08-27 | 2022-07-08 | Image processing device and image processing method |
JP2023543740A JPWO2023026706A1 (ja) | 2021-08-27 | 2022-07-08 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2021-139428 | 2021-08-27 | ||
JP2021139428 | 2021-08-27 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2023026706A1 true WO2023026706A1 (ja) | 2023-03-02 |
Family
ID=85322743
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2022/027173 WO2023026706A1 (ja) | 2021-08-27 | 2022-07-08 | 画像処理装置及び画像処理方法 |
Country Status (4)
Country | Link |
---|---|
EP (1) | EP4394697A1 (ja) |
JP (1) | JPWO2023026706A1 (ja) |
CN (1) | CN117897739A (ja) |
WO (1) | WO2023026706A1 (ja) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012069084A (ja) * | 2010-08-27 | 2012-04-05 | Hitachi High-Technologies Corp | 重み付きテンプレートマッチング実行装置およびプログラム |
JP2012208887A (ja) * | 2011-03-30 | 2012-10-25 | Denso Corp | 対象物追跡装置および対象物追跡システム |
WO2018230016A1 (ja) * | 2017-06-13 | 2018-12-20 | 株式会社Ihi | 移動体観測方法 |
-
2022
- 2022-07-08 WO PCT/JP2022/027173 patent/WO2023026706A1/ja active Application Filing
- 2022-07-08 CN CN202280058115.9A patent/CN117897739A/zh active Pending
- 2022-07-08 JP JP2023543740A patent/JPWO2023026706A1/ja active Pending
- 2022-07-08 EP EP22860997.0A patent/EP4394697A1/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012069084A (ja) * | 2010-08-27 | 2012-04-05 | Hitachi High-Technologies Corp | 重み付きテンプレートマッチング実行装置およびプログラム |
JP2012208887A (ja) * | 2011-03-30 | 2012-10-25 | Denso Corp | 対象物追跡装置および対象物追跡システム |
WO2018230016A1 (ja) * | 2017-06-13 | 2018-12-20 | 株式会社Ihi | 移動体観測方法 |
Non-Patent Citations (3)
Title |
---|
SAITO, M.HASHIMOTO, M.: "Robust Image Matching based on Estimation of Disturbed Pixels using Statistical Analysis of Intensity Variation", VISION ENGINEERING WORKSHOP (VIEW2013 |
SAITO, M.HASHIMOTO, M.: "Robust Image Matching for Illumination Variation using Stable Pixels based on Co-occurrence Analysis", INFORMATION PROCESSING SOCIETY OF JAPAN SPECIAL INTEREST GROUP TECHNICAL REPORT, vol. 2012, no. 17, 2012 |
SHINOHARA, N.HASHIMOTO, M.: "An Image-Matching Method using Template Updating based on Statistical Prediction of Outlier-Pixel", JOURNAL OF THE JAPAN SOCIETY FOR PRECISION ENGINEERING, vol. 83, 2017, pages 11178 - 1183 |
Also Published As
Publication number | Publication date |
---|---|
EP4394697A1 (en) | 2024-07-03 |
CN117897739A (zh) | 2024-04-16 |
JPWO2023026706A1 (ja) | 2023-03-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP4855556B1 (ja) | 動体検出装置、動体検出方法、動体検出プログラム、動体追跡装置、動体追跡方法及び動体追跡プログラム | |
CN103703472B (zh) | 用于确定视频中的对象的部位及相关属性的方法和*** | |
US8306282B2 (en) | Hierarchical face recognition training method and hierarchical face recognition method thereof | |
JP5500024B2 (ja) | 画像認識方法及び装置並びにプログラム | |
JP6188976B2 (ja) | 画像に含まれるテキストを検出する方法、装置及びコンピュータ読み取り可能な記録媒体 | |
JP2009265827A (ja) | 対象物検出装置及び方法、対象物検出システム、プログラム | |
US20150279054A1 (en) | Image retrieval apparatus and image retrieval method | |
JP2007122218A (ja) | 画像分析装置 | |
US8917957B2 (en) | Apparatus for adding data to editing target data and displaying data | |
TWI776176B (zh) | 手部作業動作評分裝置、方法及電腦可讀取存儲介質 | |
US11334759B2 (en) | Information processing apparatus, information processing method, and medium | |
US20230112904A1 (en) | Dynamic detection and recognition of media subjects | |
JP2011210054A (ja) | 対象物検知装置及びその学習装置 | |
JPWO2015136709A1 (ja) | 画像処理装置、画像センサ、画像処理方法 | |
US11373326B2 (en) | Information processing apparatus, information processing method and storage medium | |
JP4495606B2 (ja) | 色識別装置及び色識別プログラム | |
US20180144074A1 (en) | Retrieving apparatus, display device, and retrieving method | |
CN117474918B (zh) | 异常检测方法和装置、电子设备以及存储介质 | |
WO2023026706A1 (ja) | 画像処理装置及び画像処理方法 | |
Orjuela et al. | Evaluation of the wear label description in carpets by using local binary pattern techniques | |
JP2008140093A (ja) | 異常事象抽出装置、異常事象抽出方法、この方法のプログラム及びこのプログラムを記録した記録媒体 | |
KR101521136B1 (ko) | 얼굴 인식 방법 및 얼굴 인식 장치 | |
JP2017175466A (ja) | 映像処理装置、映像処理システムおよび制御方法 | |
KR102538804B1 (ko) | 인공 지능을 이용한 얼굴 랜드마크 검출 장치 및 방법 | |
KR102290857B1 (ko) | 채널상태정보를 이용한 인공지능 기반의 스마트 사용자 검출 방법 및 장치 |
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: 22860997 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2023543740 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 202280058115.9 Country of ref document: CN |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2022860997 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 2022860997 Country of ref document: EP Effective date: 20240327 |