WO2010016103A1 - Image collating unit and image collating method - Google Patents

Image collating unit and image collating method Download PDF

Info

Publication number
WO2010016103A1
WO2010016103A1 PCT/JP2008/063976 JP2008063976W WO2010016103A1 WO 2010016103 A1 WO2010016103 A1 WO 2010016103A1 JP 2008063976 W JP2008063976 W JP 2008063976W WO 2010016103 A1 WO2010016103 A1 WO 2010016103A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
sharpness
registered
input
information
Prior art date
Application number
PCT/JP2008/063976
Other languages
French (fr)
Japanese (ja)
Inventor
河田　耕三
Original Assignee
グローリー株式会社
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 グローリー株式会社 filed Critical グローリー株式会社
Priority to JP2010523662A priority Critical patent/JP5395795B2/en
Priority to PCT/JP2008/063976 priority patent/WO2010016103A1/en
Publication of WO2010016103A1 publication Critical patent/WO2010016103A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/73—Deblurring; Sharpening
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/34—Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands

Definitions

  • the present invention relates to an image collation apparatus and an image collation method, and more particularly to an image collation apparatus and an image collation method for collating a stored registered image with an input image that has been input.
  • an image pickup device such as a CCD (Charge-Coupled Device) has been provided, and an input is performed by collating a registered image previously captured and stored with the image-capturing device with an input image captured and input later with the image-capturing device.
  • CCD Charge-Coupled Device
  • image data of the front and back surfaces of a plurality of cards used in an ATM is stored and registered in advance.
  • the front and back surfaces of the card are imaged by the CCD module to obtain image data.
  • the image data obtained at this time and the image data of the card registered in advance are obtained.
  • a card identification device that verifies and authenticates a card inserted into a card insertion slot. JP 2007-122503 A
  • the conventional image matching device if both the image quality of the registered image and the image quality of the input image are clear, it is possible to accurately determine that the input image and the registered image are the same when the imaging target is the same. However, when the image quality of one of the registered image and the input image is not clear, it may be determined that the input image and the registered image are not the same even if the imaging target is the same. That is, in this case, the matching accuracy of the image is lowered.
  • the contour of the registered image or input image is detected and used as a feature of the imaging target, or the density change in the registered image or input image If the sharpness of the registered image and the input image differ greatly, the detected image will have different characteristics even if the image capture target is the same. Therefore, it is determined that the imaging target of the input image and the registered image are not the same.
  • the input image is often imaged under relatively poor imaging conditions, and the sharpness often varies. . Also, depending on the image collating device, the sharpness of the registered image may be extremely lower than the sharpness of the input image.
  • the conventional image matching device when the sharpness of the input image and the sharpness of the registered image are extremely different from each other, the conventional image matching device must not input images and registered images obtained by capturing the same imaging target. There is a problem that an erroneous determination may occur, and the accuracy of image collation is reduced.
  • An object of the present invention is to provide an image collation apparatus and an image collation method capable of reducing the degree of decrease in image collation accuracy.
  • the invention of claim 1 is an image collating apparatus, which acquires an image information storage unit storing a registered image and an input image to be collated with the registered image.
  • Image acquisition means sharpness detection means for detecting the sharpness of the image
  • the sharpness detection means detects the sharpness of the registered image and the input image
  • the sharpness adjustment means detects at least one of the images.
  • Sharpness difference reducing means for adjusting the sharpness to reduce the difference in sharpness between the registered image and the input image, and the registered image in which the difference in sharpness is reduced by the sharpness difference reducing means and the
  • the image collating apparatus has image collating means for collating with the input image.
  • the invention of claim 2 is an image collating device, wherein a sharpness detecting means for detecting a sharpness of an image, a sharpness adjusting means for adjusting a sharpness detected by the sharpness detecting means, An image information storage unit that stores an image adjusted by the sharpness adjustment unit as a registered image so that the sharpness detected by the sharpness detection unit approaches a predetermined value, and an input image to be compared with the registered image is acquired.
  • a sharpness difference reduction that reduces the sharpness difference between the input image and the registered image by adjusting the sharpness of the input image acquired by the image acquisition means and the image acquisition means to approach the predetermined value
  • an image collating device that collates the input image whose sharpness difference is reduced by the sharpness difference reducing unit and the registered image.
  • the invention of claim 3 is an image collating device, wherein a sharpness detecting means for detecting a sharpness of an image, a sharpness adjusting means for adjusting a sharpness detected by the sharpness detecting means, Image information storage means storing registered feature information obtained by extracting features from the registered image adjusted by the sharpness adjustment means so that the sharpness detected by the sharpness detection means approaches a predetermined value, and collation with the registered image
  • An image acquisition means for acquiring the input image, and adjusting the sharpness of the input image acquired by the image acquisition means so as to approach the predetermined value, and the difference in the sharpness between the input image and the registered image Sharpness difference reducing means for reducing the image quality, and input feature information obtained by extracting features of the input image whose sharpness has been adjusted by the sharpness difference reduction means and the registered feature information
  • an image matching device and an image matching unit for matching between the input image and the registered image.
  • the image collating means collates the registered image with the input image, and calculates an similarity between the registered image and the input image.
  • the invention of claim 5 further includes a collation target selection unit that selects information corresponding to a registered image to be collated with the input image from the image information storage unit, and the image collation unit includes the collation target selection unit.
  • the sharpness difference reducing means is configured to sharpen the input image and the registered image when a difference in sharpness between the input image and the registered image is larger than a predetermined difference.
  • An image collation apparatus that reduces the difference in degrees was used.
  • an image collation method wherein the first image and the second image are adjusted by adjusting a sharpness of at least one of the first image and the second image. And a step of collating the first image with the sharpness difference reduced with the second image.
  • an image collation apparatus an image information storage unit that stores a registered image, an image acquisition unit that acquires an input image to be collated with the registered image, and a sharpness detection that detects the sharpness of the image.
  • the sharpness detecting means detects the sharpness of the registered image and the input image
  • the sharpness adjusting means adjusts the sharpness of at least one of the images, and the registered image and the input image Sharpness difference reducing means for reducing the difference in sharpness of the image
  • an image matching apparatus having image matching means for matching the registered image and the input image in which the difference in sharpness is reduced by the sharpness difference reducing means, and Therefore, even if the sharpness of the registered image and the sharpness of the input image are different so as to affect the matching accuracy of the image as it is, the degree of decrease in the matching accuracy of the image can be reduced. It is possible to provide an image collating apparatus.
  • an image collating apparatus wherein a sharpness detection unit that detects a sharpness of an image, a sharpness adjustment unit that adjusts a sharpness detected by the sharpness detection unit, and the sharpness An image information storing means for storing an image adjusted by the sharpness adjusting means so that the sharpness detected by the degree detecting means approaches a predetermined value, and an image for acquiring an input image to be collated with the registered image And a sharpness difference reduction unit that adjusts the sharpness of the input image acquired by the image acquisition unit so as to approach the predetermined value and reduces a difference in the sharpness of the input image and the registered image.
  • an image collating device that collates the input image whose sharpness difference has been reduced by the sharpness difference reducing unit with the registered image, so that the sharpness of the registered image and the input image And sharpness of, is intact even when different enough to affect the matching precision of the image, it is possible to provide an image collating apparatus capable of reducing the deterioration degree of the matching precision of the image.
  • the processing time required for image collation can be shortened.
  • an image collating apparatus wherein a sharpness detection unit that detects a sharpness of an image, a sharpness adjustment unit that adjusts a sharpness detected by the sharpness detection unit, and the sharpness Image information storage means for storing feature information obtained by extracting features from the registered image adjusted by the sharpness adjustment means so that the sharpness detected by the degree detection means approaches a predetermined value, and input for collating with the registered image
  • An image acquisition unit that acquires an image, and a sharpness of the input image acquired by the image acquisition unit is adjusted to be close to the predetermined value to reduce a difference in the sharpness between the input image and the registered image
  • a sharpness difference reducing unit that matches the input feature information obtained by extracting features of the input image whose sharpness has been adjusted by the sharpness difference reducing unit and the registered feature information.
  • the image collation apparatus Since the image collation apparatus has an image collation unit that collates the registered image with the registered image, the sharpness of the registered image and the sharpness of the input image are different from each other so that the collation accuracy of the image is affected as it is. Even if it exists, the image collation apparatus which can make the fall degree of the collation precision of an image small can be provided. In addition, since it is not necessary to adjust the sharpness of the registered image and extract the registered feature information at the time of collation, the processing time required for image collation can be shortened.
  • the image collating unit collates the registered image with the input image and calculates an similarity between the registered image and the input image. Even if the sharpness and the sharpness of the input image are different so as to affect the matching accuracy of the image as it is, the reduction degree of the matching accuracy of the image can be reduced, and the input image and the registered image can be reduced. The similarity can be calculated.
  • a collation target selection unit that selects information corresponding to a registered image to be collated with the input image from the image information storage unit, and the image collation unit is selected by the collation target selection unit. Since the selected information is used as an image collating device that determines whether the imaging target of the input image and the imaging target of the registered image are the same, the sharpness of the registered image and the sharpness of the input image. However, even if the image matching accuracy is different as it is, the degree of decrease in the image matching accuracy can be reduced, and whether the input image and the registered image are the same imaging target. It can be determined whether or not.
  • the sharpness difference reducing means is configured such that when the difference in sharpness between the input image and the registered image is larger than a predetermined difference, the sharpness between the input image and the registered image. Therefore, when the difference in sharpness between the registered image and the input image is small, the sharpness difference reducing means does not reduce the difference in sharpness, so that the processing time required for image matching is reduced. Can be shortened.
  • an image collating method wherein the first image and the second image are adjusted by adjusting a sharpness of at least one of the first image and the second image. Since the image collating method has the step of reducing the difference in sharpness between the first image and the step of collating the first image and the second image in which the difference in sharpness is reduced, the difference in sharpness between the two images. However, even if it is different as it is to the extent that it affects the image collation accuracy, image collation can be performed with a reduced degree of decrease in the image collation accuracy.
  • FIG. 1 is a functional block diagram showing the configuration of the image collating apparatus according to this embodiment.
  • FIG. 2 is an explanatory diagram showing the first high-pass filter and the second high-pass filter in the present embodiment.
  • FIG. 3 is an explanatory diagram showing a procedure of image processing at the time of image registration and image verification by the image verification apparatus according to the present embodiment.
  • FIG. 4 is a flowchart showing processing executed by the image matching apparatus according to the present embodiment at the time of image registration.
  • FIG. 5 is a flowchart showing processing executed by the image collating apparatus according to the present embodiment when inputting an image.
  • FIG. 6 is a flowchart showing an image matching process executed by the image matching apparatus according to the present embodiment when inputting an image.
  • FIG. 7 is a functional block diagram illustrating a configuration of an image collating apparatus according to a modification.
  • a registered image obtained by imaging a plurality of people's ears is registered in advance as a registered image, and thereafter, an input image obtained by imaging a certain person's ear and a registered image registered in advance are checked.
  • An image collating apparatus to be performed will be described as an example.
  • the imaging target of the image collated by the image collating apparatus according to the present invention is not limited to the human ear, and may be other parts of the human body or other than the human body.
  • FIG. 1 is a functional block diagram showing the configuration of the image collating apparatus according to this embodiment.
  • the image matching device 1 includes an imaging unit 2 that captures an imaging target, a registration process that stores and registers an image captured by the imaging unit 2 as a registered image, and image verification.
  • Information processing unit 3 that performs a collation process or the like for collating an image (hereinafter referred to as an “input image”) captured later to perform registration with a registered image, and registration that causes the information processing unit 3 to perform a registration process Mode selection unit 4 operated to switch between the mode and the collation mode for executing the collation processing, and various images indicating the mode selected by the mode selection unit 4 and the collation result of the input image and the registered image by the information processing unit 3 Is displayed.
  • the imaging unit 2 is configured by a CCD (Charge-Coupled Device) camera, converts an analog image signal obtained by imaging a human ear that is an imaging target into digital image information, and sends the digital image information to the information processing unit 3. Output.
  • CCD Charge-Coupled Device
  • the imaging unit 2 functions as an image acquisition unit that captures a registered image and an input image to be compared with the registered image and acquires image information regarding each image.
  • the image picked up by the image pickup unit 2 is registered as a registered image.
  • the present invention is not limited to this, and a photograph of an ear picked up for registration is read and acquired.
  • the image data of the photograph or the image data of the ear captured at another location by a digital camera or the like may be registered as a registered image in the image collating apparatus of the present embodiment.
  • the information processing unit 3 includes a microcomputer having a CPU (Central Processing Unit), a ROM (Read-Only Memory), and a RAM (Random Access Memory).
  • the ROM stores various programs executed by the CPU when the information processing unit 3 performs the registration process, the collation process, and the like.
  • the RAM is a temporary storage unit that functions as a work area when the CPU executes various programs stored in the ROM.
  • the information processing unit 3 reads and executes various programs stored in the ROM by the CPU, whereby the degree of conspicuousness of the image captured by the imaging unit 2 or the like (hereinafter referred to as “sharpness”).
  • the extraction unit 8 functions as a collation unit 8 that collates the registered image and the input image.
  • the information processing unit 3 includes an image information storage unit 10 that stores image information regarding a plurality of registered images.
  • the image information storage unit 10 is configured by an HDD (Hard disk drive). Note that the image information storage unit 10 is not limited to the HDD, and a nonvolatile memory such as a flash memory may be used as long as the storage device has a large capacity.
  • the image information storage unit 10 functions as an image information storage unit that stores registered feature information indicating features of a registered image adjusted so that the sharpness approaches a predetermined value.
  • the information processing unit 3 displays an image corresponding to the image information as a registered image. Judge that there is.
  • the information processing unit 3 first functions as the sharpness detection unit 6 and detects the sharpness of the registered image based on the input image information. In this way, the information processing unit 3 functions as a sharpness detection unit that detects the sharpness of a captured image based on the image information.
  • the information processing unit 3 functions as the sharpness adjusting unit 7 and adjusts the detected sharpness.
  • the information processing unit 3 performs a process of reducing the sharpness so as to approach the predetermined value. That is, a process of slightly blurring a registered image that is too clear is performed.
  • the information processing unit 3 does not adjust the sharpness when the detected sharpness is equal to or less than a predetermined value.
  • the process of reducing the sharpness performed by the information processing unit 3 is not a process of completely blurring the image picked up by the image pickup unit 2, but a plurality of registered images to be sequentially registered can be discriminated later as separate registered images. This is a process for reducing the sharpness of each registered image.
  • the information processing unit 3 functions as a sharpness adjusting unit that adjusts the sharpness of the registered image detected by the sharpness detecting unit so as to approach a predetermined value.
  • the information processing unit 3 functions as the feature information extraction unit 8 to extract feature information indicating the feature amount of the image based on the image information of the image whose sharpness is lowered to a predetermined value, and the extraction
  • the obtained feature information is stored in the image information storage unit 10 as image information of the registered image (registered feature information).
  • the feature information here is, for example, the position of a characteristic shape in the image and the degree of change in the shading of the image.
  • the information processing unit 3 functions as a feature information extraction unit that extracts feature information of an image from a registered image in which the difference in sharpness is reduced.
  • the information processing unit 3 also extracts the feature information of the registered image that does not require sharpness adjustment and stores it in the image information storage unit 10 as image information of the registered image.
  • the information processing unit 3 sequentially functions as the sharpness detection unit 6, the sharpness adjustment unit 7, and the feature information extraction unit 8 each time new image information is input from the imaging unit 2, and the image information storage unit In 10, image information of a plurality of registered images is registered.
  • the sharpness of the registered image detected by the sharpness detection unit 6 is adjusted so as to approach a predetermined value by the sharpness adjustment unit 7, and then the registered image is characterized from the registered image whose sharpness has been adjusted.
  • the information is extracted and stored in the image information storage unit 10, not the feature information but the image information of the registered image adjusted so that the sharpness approaches a predetermined value is stored in the image information storage unit 10.
  • the image information storage unit 10 functions as an image information storage unit that stores a registered image whose sharpness is adjusted to approach a predetermined value by the sharpness adjustment unit.
  • the information processing unit 3 displays an image corresponding to the image information as an input image. It is judged that.
  • the information processing unit 3 first functions as the sharpness detection unit 6 to detect the sharpness of the input image, and then functions as the sharpness adjustment unit 7 to adjust the detected sharpness. At this time, as in the registration mode, when the detected sharpness is higher than a predetermined value described later, the information processing unit 3 performs a process of reducing the sharpness so as to approach the predetermined value. As described above, in the present embodiment, the information processing unit 3 adjusts the sharpness of the input image detected by the sharpness detection unit so as to approach a predetermined value, and determines the sharpness of the input image and the registered image. It functions as a sharpness difference reducing means for reducing the difference.
  • the information processing unit 3 functions as the feature information extraction unit 8 to extract the feature information (input feature information) of the input image with the sharpness adjusted, as in the registration mode.
  • the information processing unit 3 extracts the feature information (input feature information) of the input image based on the image information of the input image whose sharpness is not adjusted.
  • the information processing unit 3 functions as a feature information extraction unit that extracts feature information of an image from an input image in which a difference in sharpness is reduced.
  • the information processing unit 3 functions as the matching unit 9 to compare and match the input feature information extracted from the input image with the registered feature information of the registered image stored in the image information storage unit 10. It is determined whether or not the imaging target of the input image and the imaging target of the registered image are the same.
  • the information processing unit 3 compares the input feature information indicating the feature of the input image whose sharpness has been adjusted by the sharpness difference reducing unit with the registered feature information, thereby obtaining the input image and the registered image. It functions as image collating means for collation.
  • the information processing unit 3 sequentially reads the registered feature information registered in the image information storage unit 10 for each registered image, and sequentially compares the registered feature information of the registered image with the input feature information of the input image. To do.
  • the information processing unit 3 displays an image indicating that on the display unit, while there is a matching registered image. If not, an image indicating that is displayed on the display unit.
  • the information processing unit 3 sets the sharpness of each registration image to a predetermined value. By making the adjustment to be uniform, the variation in sharpness is reduced, and the feature information of each image with the sharpness being almost uniform is registered as the registered feature information of the registered image.
  • the information processing unit 3 adjusts the sharpness of the input image so that it is substantially equal to the uniform sharpness of the registered image.
  • the input feature information of the input image after adjustment and the registered feature information of the registered image are compared, and the input image and the registered image are collated.
  • the information processing unit 3 adjusts at least one of the difference between the sharpness of the registered image and the input image detected by the sharpness detecting unit, and the sharpness between the registered image and the input image is adjusted. It functions as a sharpness difference reducing means for reducing the difference between the two.
  • FIG. 2 is an explanatory diagram showing an image processing filter used for detection of sharpness.
  • the digital image information input from the imaging unit 2 is referred to as a first high-pass filter and a second high-pass filter illustrated in FIG. 2.
  • the sharpness of each image is detected from image information of an image captured as an input image or a registered image.
  • the image information input to the information processing unit 3 is digital information in which each pixel of a captured image is represented by a plurality of binary information of 0 or 1.
  • the information processing unit 3 causes F1A and F1B as the first high-pass filter to act on the image information input from the imaging unit 2, and applies the first information to the information obtained as a result and the image information input from the imaging unit 2.
  • F2 as a high-pass filter 2 is applied, and the sharpness of the captured image is detected using information obtained as a result.
  • This first high-pass filter is called a Sobel differential operator, and is an operator that is widely used to extract image edges.
  • (output 1)
  • * means convolution
  • means the absolute value of X.
  • the second high-pass filter is called a Laplacian operator, which is also an operator that is widely used for extracting image edges.
  • (output 2)
  • sharpness detected by the information processing unit 3 is (sharpness)
  • the information processing unit 3 When functioning as the sharpness adjusting unit 7, the information processing unit 3 performs a process of smoothing the image information, thereby setting the sharpness of an image whose sharpness is higher than a predetermined value as a predetermined value.
  • the information processing unit 3 first performs Fourier transform on the image information input from the imaging unit 2 and multiplies the information on the frequency W obtained as a result by a multiplier M (W).
  • W SQRT (U ⁇ U + V ⁇ V)
  • M (W) 1 ⁇ ⁇ (S ⁇ S0) ⁇ W.
  • ⁇ is a control parameter.
  • the information processing unit 3 smoothes the image information by removing a predetermined amount of harmonic components from the image information input from the imaging unit 2, and sets the sharpness of an image whose sharpness is higher than a predetermined value to a predetermined value. Value.
  • the information processing unit 3 When functioning as the feature information extraction unit 8, the information processing unit 3 first divides the image corresponding to the image information whose sharpness has been adjusted into M ⁇ N blocks.
  • the information processing unit 3 sequentially applies F Gabor filters to information corresponding to each block divided into M ⁇ N blocks, sets P ⁇ Q sample points in each block, F data at the sample point are extracted.
  • the Gabor filter used here is a filter that is widely used for extracting only the contour line of an image, and is a filter that extracts only a contour line whose direction matches the Gabor function. Using the F pieces of data thus extracted, a P ⁇ Q ⁇ F-dimensional feature amount is acquired, and this feature amount is extracted as feature information of the image.
  • the information processing unit 3 stores and registers the feature information (registered feature information) of the registered image extracted in this manner in the image information storage unit 10 for each registered image.
  • the feature information (input feature information) of the input image is compared with the registered feature information of the registered image registered in the image information storage unit 10, and the input image and the registered image are registered. Match the image.
  • FIG. 3 is an explanatory diagram showing an image processing procedure of the image collation apparatus at the time of image registration and image collation.
  • the predetermined sharpness predetermined value is 0.21.
  • the numerical value of the sharpness is a numerical value that increases as the image becomes clearer.
  • the image collation apparatus 1 of the present embodiment a plurality of registered images are stored and registered in advance, and thereafter, the input image is input and the input image and the registered image are collated.
  • an image processing procedure performed on an image captured by the image matching device 1 at the time of image registration will be described.
  • the image display apparatus 1 captures an image of a person's ear that is an imaging target, as shown in FIG. Thereafter, the image matching device 1 detects the sharpness of the image based on the image information of the captured image.
  • the image matching apparatus 1 lowers the sharpness of the image to 0.21 because the detected sharpness is higher than a predetermined value of 0.21. Perform sharpness adjustment processing. On the other hand, the image matching device 1 does not adjust the sharpness when the sharpness of the captured image is 0.21 or less.
  • the image matching device 1 performs a process of extracting feature information of the image from the image information after the sharpness adjustment.
  • the image matching device 1 performs a process of extracting feature information of the image from the image information that has not been subjected to the sharpness adjustment.
  • the image matching device 1 performs the above-described image processing on each of the plurality of images captured at the time of image registration, and then stores the feature information of each image in the image information storage unit 10 as registered feature information of the registered image.
  • the image display device 1 captures an image of a human ear that is an imaging target when inputting an image. Thereafter, the image matching device 1 detects the sharpness of the image based on the image information of the captured image.
  • the image matching device 1 performs processing for adjusting the sharpness of the captured image because the detected sharpness is equal to or less than a predetermined value of 0.21. Not performed.
  • the image collating apparatus 1 performs a process of reducing the sharpness of the image to a predetermined value of 0.21.
  • the image matching device 1 performs a process of extracting feature information of the image from the image information on which the sharpness is not adjusted. Further, when the sharpness is adjusted, the image matching device 1 performs a process of extracting feature information of the image from the image information after the sharpness is adjusted.
  • the image collation apparatus 1 extracts the feature information of the image captured at the time of inputting the image in this way, the input feature information that is the feature information of the image and a plurality of registered features that are already stored in the image information storage unit 10. Processing for sequentially collating information is performed, and the collation result is output to the display unit 5.
  • FIG. 4 is a flowchart showing processing executed by the image matching device 1 at the time of image registration
  • FIG. 5 is a flowchart showing processing executed by the image matching device 1 at the time of image input
  • FIG. 4 is a flowchart illustrating an image matching process that is sometimes executed by the image matching apparatus 1.
  • the image collation apparatus 1 determines whether or not the imaging unit 2 has detected an ear of a person who is an imaging target (step S101) and determines that an imaging target has been detected (step S101: Yes). Then, the image capturing target is imaged, and image information related to the captured registered image is output to the information processing unit 3, and then the process proceeds to step S102. On the other hand, when the imaging unit 2 does not detect the imaging target (step S101: No), the process ends.
  • step S ⁇ b> 102 in the image collation device 1, the CPU of the information processing unit 3 reads a predetermined program from the ROM and executes it, thereby sharpening the registered image based on the image information of the registered image input from the imaging unit 2. A process for detecting the degree is performed, and then the process proceeds to step S103.
  • step S103 the image collation apparatus 1 determines whether the sharpness of the registered image detected by the information processing unit 3 in step S102 is higher than a predetermined value of 0.21, and is higher than 0.21. If it is determined (step S103: Yes), the process proceeds to step S104. If it is determined that the value is 0.21 or less (step S103: No), the process proceeds to step S105.
  • step S104 in the image collation device 1, the information processing unit 3 performs a process of reducing the sharpness detected in step S102 to a predetermined value of 0.21, and then the process proceeds to step S105.
  • step S105 if the sharpness is adjusted in step S104, the information processing section 3 extracts the feature information (registered feature information) of the registered image from the image information after the sharpness adjustment.
  • the process extracts the registered image feature information (registered feature information) from the image information input from the imaging unit 2.
  • the information processing unit 3 performs processing for storing the registered feature information extracted in step S105 in the image information storage unit 10 as the feature amount of the registered image (step S106), and ends the processing.
  • step S101 to step S106 is sequentially executed, and the image information storage unit In FIG. 10, image information of a plurality of registered images is registered.
  • the registered feature information of a plurality of registered images is stored in the image information storage unit 10, but not the registered feature information but the sharpness is brought closer to a predetermined value (here, 0.21).
  • the image information of the registered image that has been adjusted may be stored.
  • the image collation apparatus 1 determines whether or not the imaging unit 2 has detected an ear of a person who is an imaging target (step S201), and determines that an imaging target has been detected (step S201: Yes). Then, the image capturing target is imaged, and image information related to the captured input image is output to the information processing unit 3, and then the process proceeds to step S202. On the other hand, when the imaging unit 2 does not detect the imaging target (step S201: No), the process ends.
  • step S202 in the image collation device 1, the CPU of the information processing unit 3 reads a predetermined program from the ROM and executes it, thereby sharpening the input image based on the image information of the input image input from the imaging unit 2. A process for detecting the degree is performed, and then the process proceeds to step S203.
  • step S203 the image matching device 1 determines whether or not the sharpness of the input image detected by the information processing unit 3 in step S202 is higher than a predetermined value of 0.21, and is higher than 0.21. If it is determined (step S203: Yes), the process proceeds to step S204. If it is determined that the value is 0.21 or less (step S203: No), the process proceeds to step S205.
  • step S204 in the image collation device 1, the information processing unit 3 performs a process of lowering the sharpness detected in step S202 to a predetermined value of 0.21, and then the process proceeds to step S205.
  • step S205 if the sharpness adjustment is performed in step S204, the information processing unit 3 extracts the feature information (input feature information) of the input image from the image information after the sharpness adjustment.
  • the processing is performed and the sharpness is not adjusted, the processing is performed to extract the feature information (input feature information) of the input image from the image information input from the imaging unit 2, and then the processing proceeds to step S206. Transfer.
  • step S206 in the image collation apparatus 1, the information processing unit 3 performs image collation processing for comparing and collating the input image and the registered image, and then ends the processing.
  • the image collating process will be specifically described with reference to FIG.
  • the information processing unit 3 performs a process of setting “1” as the variable Z in a predetermined amount area in the RAM (step Thereafter, the process proceeds to step S302.
  • variable Z set by the information processing unit 3 is an identification number sequentially assigned by the information processing unit 3 when each registered image is registered in the image information storage unit 10, and is used for distinguishing each registered image. This is the identification information.
  • the information processing unit 3 reads the registered feature information of the registered image from the image information storage unit 10, the information processing unit 3 reads the registered feature information of the Zth registered image.
  • step S302 in the image matching device 1, after the information processing unit 3 reads the registered feature information of the Zth registered image registered in the image information storage unit 10, the input feature information of the input image and the registration of the registered image are registered. A process of comparing and collating with feature information is performed (step S303), and then the process proceeds to step S304.
  • the input feature information and the registered feature information are compared and collated to collate the input image and the registered image.
  • the image information storage unit 10 is not registered feature information but sharp.
  • the image information of the registered image after adjusting the degree to approach the predetermined value is stored, the image information of the registered image after adjusting the sharpness stored in the image information storage unit 10 and Similarly, the input image and the registered image are collated by comparing and collating with the image information of the input image after adjusting the sharpness to be close to a predetermined value.
  • step S304 the image matching apparatus 1 determines whether or not the input feature information of the input image compared and matched by the information processing unit 3 in step S303 matches the registered feature information of the registered image. If so (step S304: Yes), the process proceeds to step S307, and a matching result display process for displaying an image indicating that the input feature information of the input image matches the registered feature information of the registered image on the display unit 5 is performed. After that, the process ends.
  • step S304 determines that the input feature information of the input image compared and collated in step S303 does not match the registered feature information of the registered image (step S304: No).
  • step S305 A process of adding “1” to the variable Z is performed, and then the process proceeds to step S306.
  • step S306 in the image collation device 1, whether or not the variable Z has reached the maximum value Zmax (the number of registered images registered) as a result of the information processing unit 3 adding “1” to the variable Z in step S305. If it is determined that Zmax has not been reached (step S306: No), the process proceeds to step S302. If it is determined that Zmax has been reached (step S306: Yes), the process proceeds to step S307. .
  • step S307 in the image collation device 1, when the information processing unit 3 determines that the variable Z has reached Zmax in step S306, the registered image having the registered feature information that matches the input feature information of the input image.
  • a collation result display process for displaying an image indicating that no is present on the display unit 5 is performed, and then the process ends.
  • it is configured to determine whether or not the input feature information of the input image matches the registered feature information of the registered image in step S304, but the determination in step S304 is not performed.
  • it may be configured to determine whether or not the input feature information of the input image matches the registered feature information of the registered image.
  • the matching result between the input image and the registered image that is, the input image and the registered image match, or the registered image that matches the input image is displayed.
  • the information processing unit 3 sequentially collates the input image with all the registered images, calculates the similarity between the input image and each registered image, and inputs
  • a plurality of registered images may be arranged and displayed on the display unit 5 in descending order of similarity to images. Further, in such a configuration, only registered images whose similarity to the input image exceeds a predetermined threshold may be displayed on the display unit 5 in descending order of similarity.
  • the image collation apparatus 1 detects the sharpness of each image for each of a plurality of images captured in advance for registration, and when the sharpness is higher than a predetermined value, the sharpness is determined. By lowering the degree to a predetermined value, the sharpness of each captured image is made uniform so as to be substantially the same, and then the registered feature information of the image is stored in advance in the image information storage 10 as the feature amount of the registered image. Let me register.
  • the image matching device 1 detects the sharpness of the image captured for input, and reduces the sharpness to a predetermined value only when the sharpness is higher than the predetermined value.
  • the input feature information of the image is extracted as the feature quantity of the input image, and the input feature information of the input image is compared with the registered feature information of the registered image, and the input image and the registered image are compared. ing.
  • this image collation apparatus 1 when the input image and the registered image are collated, the sharpness of each image is substantially the same. Therefore, the sharpness of the registered image and the sharpness of the input image are captured at the time of imaging. However, even if it is different so as to affect the collation accuracy of the image, the collation accuracy of the image is not lowered.
  • the image information storage unit 10 includes Image information corresponding to one registered image is registered, a plurality of input images are input to the image collating apparatus 1, and the same imaging target as that of the registered image is captured from the plurality of input images. It is also possible to determine the input image.
  • the collation between the input image and the registered image is repeatedly performed until a registered image that matches the input feature information and the feature information of the input image is found.
  • a registered image to be collated with the input image may be selected in advance, and the image collating apparatus may be configured to collate the registered image with the input image.
  • FIG. 7 is a functional block diagram illustrating an image collating apparatus according to a modification.
  • the same reference numerals are given to parts having the same functions as those of the image collating apparatus shown in FIG. 1.
  • the image collation apparatus 1a according to the present modification is similar to the image collation apparatus 1 shown in FIG. 1 with respect to its basic configuration, but includes a registered image selection unit 11. Different from the image collating apparatus 1 shown in FIG.
  • the image collation apparatus 1a includes an imaging unit 2, an information processing unit 3, a mode selection unit 4, and a display unit 5, similarly to the image collation device 1 shown in FIG.
  • the information processing unit 3 executes a predetermined program stored in the ROM by the internal CPU, so that the sharpness detection unit 6, the sharpness adjustment unit 7, the feature information extraction unit 8, the collation unit 9, and the like It functions.
  • the information processing unit 3 also includes an image information storage unit 10a that stores image information of a plurality of registered images.
  • the image information storage unit 10 of the image collating apparatus 1 shown in FIG. 1 only the feature information of a plurality of registered images is stored, but the image information storing unit 10a of the image collating apparatus 1a according to this modification is stored. All the information related to the image captured for registration or the image for registration captured in advance at another location is stored as image information of the registered image for each image.
  • the image information stored in the image information storage unit 10 is image information of the registered image itself that has not been subjected to sharpness adjustment.
  • this image collation apparatus 1a images an input image, it is the image information of the specific registration image collated with an input image from the image information of the some registration image memorize
  • a registered image selection unit 11 operated to select the image.
  • the registered image selection unit 11 in this modification functions as a collation target selection unit that selects information corresponding to a registered image to be collated with the input image from the image information storage unit.
  • this image collation apparatus 1a when a registration image is captured in advance, a unique identification number (for example, the variable Z) is assigned to each captured registered image, and the identification number is displayed on the display unit 5.
  • a unique identification number for example, the variable Z
  • the registered image selection unit 11 is operated to select and input the identification number displayed on the display unit 5 when registering the registered image.
  • the information processing unit 3 reads out the image information of the registered image corresponding to the input identification number as a target for collation of the input image from the image information storage unit 10a. Based on this, the sharpness of the registered image is detected, the sharpness is adjusted, and the feature information is extracted.
  • the feature information of the input image is compared with the feature information of the selected registered image, and the input image and the registered image are compared. To determine whether or not the imaging target of the input image is the same as the imaging target of the registered image.
  • the image collating apparatus 1a is characterized by selecting a registered image to be collated with the input image in advance and adjusting the sharpness of the image information of the registered image as necessary. Information is extracted and compared with the feature information of the input image.
  • the image collating apparatus 1a it is not necessary to perform the image collating process repeatedly until a registered image whose feature information matches the input image is found, so that the processing time required for the image collating process can be shortened. it can.
  • the sharpness of the selected registered image is adjusted to a predetermined value, and the sharpness of the input image is also set to a predetermined value as necessary. Therefore, when the input image and the registered image are collated, the sharpness of the input image and the registered image is substantially the same, and the sharpness of the registered image and the sharpness of the input image are obtained at the time of imaging. Even if they differ to such an extent that they affect the collation accuracy of the image, the collation accuracy of the image does not decrease.
  • the image information storage unit 10a of the image collating apparatus 1a includes all of the registered images input from the imaging unit 2 to the information processing unit 3 when a registration image is captured. Therefore, when collating the input image with the registered image, the sharpness of the registered image can be adjusted to be higher than a predetermined value.
  • the image information storage unit 10a does not store the feature information of the image with the sharpness adjusted uniformly as in the image information storage unit 10 shown in FIG. 1, but the image information before adjusting the sharpness. Is remembered. Therefore, the sharpness of the registered image can be changed relatively flexibly.
  • the information processing unit 3 The feature information of the input image can be compared with the feature information of the registered image in a state where the sharpness of the registered image is kept relatively high without lowering to a predetermined value.
  • the input image having a relatively high sharpness and the registered image can be collated, so that the accuracy of image collation can be further improved.
  • the sharpness of the registered image can be matched with the sharpness of the input image. This can also reduce the difference between the sharpnesses of the input image and the registered image when collating, thereby improving the accuracy of image collation.
  • the image collating apparatus 1a uses the image information of the selected registered image to display the registered image. It is also possible to perform image processing that emphasizes the edge portion, extract feature information after approaching the sharpness of the input image, and compare the feature information with the feature information of the input image. Thereby, the precision of image collation can be further improved.
  • the imaging unit 2 processes all image information related to the input image. Since the image is input to the unit 3, when the sharpness of the input image is lower than a predetermined value, the edge portion of the input image is used by using the image information of the input image input from the imaging unit 2. It is also possible to perform image processing for emphasizing the image, extract the feature information after bringing the sharpness of the input image close to a predetermined value, and compare the feature information with the feature information of the selected registered image. Thereby, the accuracy of image collation can be further improved.
  • the sharpness is calculated.
  • the method of reducing the sharpness difference by the sharpness difference reducing means of the present invention is not limited to this, and the difference between the sharpness of the input image and the sharpness of the registered image is calculated. Any method may be used as long as the method is reduced.
  • the sharpness may be adjusted to be substantially the same as the predetermined value.
  • the sharpness of the image with the higher sharpness may be reduced in accordance with the image with the lower sharpness of the input image and the registered image.
  • the sharpness of the lower sharpness image may be increased in accordance with the higher sharpness image of the two images.
  • the sharpness difference reducing means is configured to reduce the difference in sharpness between the input image and the registered image when the difference between the sharpness of the input image and the registered image is larger than a predetermined difference. May be.
  • the image information of the image captured by the image capturing unit 2 is stored and registered in the image information storage unit 10 as a registered image, but the registered image registered in the image information storage unit 10 is The image need not necessarily be an image captured by the imaging unit 2.
  • image data obtained by reading a photograph of the registration target with a scanner or image data obtained by capturing the registration target with a digital camera is directly input to the image processing apparatus 1.
  • the image information storage unit 10 may be registered.
  • the input image is configured such that image data obtained by reading a photograph of the registration target with a scanner or image data obtained by capturing the registration target with a digital camera is directly input to the image processing apparatus 1 as an input image. May be.
  • the image matching method between the input image and the registered image by the image matching device 1 has been described. It is not limited to the image collation method for collating the registered image, and the first image is adjusted by adjusting the sharpness of at least one of the two arbitrary images of the first image and the second image. It is characterized in that the difference in sharpness between the image and the second image is reduced, and the first image and the second image in which the difference in sharpness is reduced are collated.
  • the image collating method of the present invention even if the difference in sharpness between two arbitrary images, ie, the first image and the second image, is different enough to affect the collation accuracy of the image.
  • the degree of decrease in collation accuracy can be reduced, and it can be accurately determined whether or not the imaging targets of any two images of the first image and the second image are the same.
  • a flexible disk FD
  • a CD-ROM Compact Disc
  • MO disk Compact Disc
  • DVD disk a magneto-optical disk
  • an IC inserted into a computer.
  • "Portable physical medium” such as a card, or "fixed physical medium” such as a hard disk drive (HDD) provided inside or outside the computer, and further via a public line, the Internet, LAN, WAN, etc.
  • Each program may be stored in “another computer (or server)” connected to the computer, and the computer may read and execute each program from these.
  • each component of each illustrated apparatus is functionally conceptual and does not necessarily need to be physically configured as illustrated.
  • the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured.
  • all or any part of each processing function performed in each device may be realized by a CPU and a program analyzed and executed by the CPU, or may be realized as hardware by wired logic.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Human Computer Interaction (AREA)
  • Image Analysis (AREA)
  • Collating Specific Patterns (AREA)
  • Image Processing (AREA)

Abstract

The collating accuracy of an image collating unit for collating the registered image and the input image is improved. The image collating unit for collating the input image and the registered image has an image information storage means to store registered images, an image capture means to capture the input image to be collated with the registered image, a sharpness degree detection means to detect the sharpness degree of the image, a sharpness degree difference reduction means for adjusting at least one of the sharpness degrees of the registered image and the input image that are detected by the sharpness degree detection means to reduce a difference between the sharpness degrees of the registered image and the input image, and an image collating means to collate the registered image and the input image where their difference in sharpness degree is reduced by the sharpness degree difference reduction means.

Description

画像照合装置及び画像照合方法Image collation apparatus and image collation method
　本発明は、画像照合装置及び画像照合方法に関するものであり、特に、記憶した登録画像と、入力された入力画像とを照合する画像照合装置及び画像照合方法に関するものである。 The present invention relates to an image collation apparatus and an image collation method, and more particularly to an image collation apparatus and an image collation method for collating a stored registered image with an input image that has been input.
　従来より、ＣＣＤ（Charge Coupled Device）等の撮像装置を備え、この撮像装置で予め撮像して記憶した登録画像と、後に撮像装置で撮像して入力された入力画像とを照合することによって、入力画像の撮像対象と登録画像の撮像対象とが同一であるか否かを判別する画像照合装置が知られている。 Conventionally, an image pickup device such as a CCD (Charge-Coupled Device) has been provided, and an input is performed by collating a registered image previously captured and stored with the image-capturing device with an input image captured and input later with the image-capturing device. 2. Description of the Related Art An image matching device that determines whether an image capturing target and a registered image capturing target are the same is known.
　この画像照合装置として、たとえば、特許文献１には、ＡＴＭ（Automatic Teller Machine：現金自動預払機）で使用される複数のカードの表面及び裏面の画像データを予め記憶して登録しておき、ＡＴＭのカード挿入口にカードが挿入された場合に、そのカードの表面及び裏面をＣＣＤモジュールで撮像して画像データを取得し、このとき取得した画像データと予め登録していたカードの画像データとを照合して、カード挿入口に挿入されたカードの真偽判定を行うカード識別装置が記載されている。
特開２００７－１２２５０３号公報 As this image collation device, for example, in Patent Document 1, image data of the front and back surfaces of a plurality of cards used in an ATM (Automatic Teller Machine) is stored and registered in advance. When the card is inserted into the card insertion slot, the front and back surfaces of the card are imaged by the CCD module to obtain image data. The image data obtained at this time and the image data of the card registered in advance are obtained. There is described a card identification device that verifies and authenticates a card inserted into a card insertion slot.
JP 2007-122503 A
　しかしながら、上記した特許文献１をはじめとする従来の画像照合装置では、登録画像の撮像対象と入力画像の撮像対象とが同一であっても、入力画像と登録画像とが同一でないと判断される場合があったため、画像の照合精度をさらに向上させる必要があった。 However, in the conventional image collation apparatus including the above-described Patent Document 1, even when the imaging target of the registered image and the imaging target of the input image are the same, it is determined that the input image and the registered image are not the same. Since there were cases, it was necessary to further improve the accuracy of image collation.
　すなわち、従来の画像照合装置では、登録画像の画質と入力画像の画質とが共に鮮明であれば、撮像対象が同一の場合、入力画像と登録画像とが同一であると正確に判断することができるが、登録画像と入力画像とのうち、いずれか一方の画質が鮮明でない場合には、撮像対象が同一であっても、入力画像と登録画像とが同一でないと判断されることがある。つまり、この場合画像の照合精度が低下してしまう。 That is, in the conventional image matching device, if both the image quality of the registered image and the image quality of the input image are clear, it is possible to accurately determine that the input image and the registered image are the same when the imaging target is the same. However, when the image quality of one of the registered image and the input image is not clear, it may be determined that the input image and the registered image are not the same even if the imaging target is the same. That is, in this case, the matching accuracy of the image is lowered.
　特に、登録画像や入力画像の輪郭の際立ち度合い（以下、「鮮鋭度」という。）が極端に異なる場合には、撮像対象が同一であっても、入力画像及び登録画像の撮像対象が同一でないと判断されることが多々ある。 In particular, when the degree of conspicuousness of the contours of the registered image and the input image (hereinafter referred to as “sharpness”) is extremely different, even if the imaging target is the same, the imaging target of the input image and the registered image is not the same. It is often judged.
　たとえば、登録画像や入力画像を構成する各画素間の濃度の変化に基づいて、登録画像や入力画像の輪郭を検出して撮像対象物の特徴とする場合や、登録画像や入力画像における濃度変化の状態を検出して撮像対象の特徴とする場合は、登録画像と入力画像との鮮鋭度が大きく異なっていると、撮像対象が同一であっても、検出した各画像の特徴が異なってしまい、入力画像及び登録画像の撮像対象が同一でないと判断されてしまう。 For example, based on a change in density between pixels constituting a registered image or input image, the contour of the registered image or input image is detected and used as a feature of the imaging target, or the density change in the registered image or input image If the sharpness of the registered image and the input image differ greatly, the detected image will have different characteristics even if the image capture target is the same. Therefore, it is determined that the imaging target of the input image and the registered image are not the same.
　登録画像や入力画像の鮮鋭度は、それら各画像を撮像するときの撮像条件の違いによってそれぞれ異なるものである。通常、登録画像の撮像では、撮像に要する時間を短縮することよりも、正確に撮像対象を撮像することに重点が置かれる。そのため、登録画像は、一定の好適な撮像条件のもとで撮像されることが多く、その鮮鋭度が比較的高く保たれていることが多い。 * The sharpness of the registered image and the input image varies depending on the image capturing conditions when capturing each image. Usually, in capturing a registered image, an emphasis is placed on accurately capturing an imaged object rather than shortening the time required for imaging. For this reason, registered images are often imaged under certain suitable imaging conditions, and their sharpness is often kept relatively high.
　これに対して、入力画像は、認証に要する時間を短縮することに重点が置かれるので、比較的悪い撮像条件のもとで撮像されることも多く、その鮮鋭度にバラツキが生じることが多い。また、画像照合装置によっては、登録画像の鮮鋭度の方が入力画像の鮮鋭度よりも極端に低い場合もありえる。 On the other hand, since an emphasis is placed on shortening the time required for authentication, the input image is often imaged under relatively poor imaging conditions, and the sharpness often varies. . Also, depending on the image collating device, the sharpness of the registered image may be extremely lower than the sharpness of the input image.
　このように、従来の画像照合装置は、入力画像の鮮鋭度と登録画像の鮮鋭度とに極端に異なっていた場合、同一の撮像対象を撮像した入力画像と登録画像とを同一の画像でないと誤判断することがあり、画像の照合精度が低下するという課題があった。 As described above, when the sharpness of the input image and the sharpness of the registered image are extremely different from each other, the conventional image matching device must not input images and registered images obtained by capturing the same imaging target. There is a problem that an erroneous determination may occur, and the accuracy of image collation is reduced.
　そこで、本発明は、上述した従来の課題を解決するべく、登録画像の鮮鋭度と入力画像の鮮鋭度とが、そのままでは画像の照合精度に影響する程に異なっていた場合であっても、画像の照合精度の低下度合いを小さくすることができる画像照合装置及び画像照合方法を提供することを目的とする。 Therefore, in order to solve the above-described conventional problems, the present invention, even if the sharpness of the registered image and the sharpness of the input image are different so as to affect the matching accuracy of the image as it is, An object of the present invention is to provide an image collation apparatus and an image collation method capable of reducing the degree of decrease in image collation accuracy.
　上述した課題を解決し、目的を達成するために、請求項１の発明は、画像照合装置であって、登録画像を記憶した画像情報記憶手段と、前記登録画像と照合する入力画像を取得する画像取得手段と、画像の鮮鋭度を検出する鮮鋭度検出手段と、前記鮮鋭度検出手段により前記登録画像及び前記入力画像の鮮鋭度を検出し、前記鮮鋭度調整手段により、少なくとも一方の画像の鮮鋭度を調整して、前記登録画像と前記入力画像との鮮鋭度の差を低減する鮮鋭度差低減手段と、前記鮮鋭度差低減手段により鮮鋭度の差が低減された前記登録画像と前記入力画像とを照合する画像照合手段と有する画像照合装置とした。 In order to solve the above-described problems and achieve the object, the invention of claim 1 is an image collating apparatus, which acquires an image information storage unit storing a registered image and an input image to be collated with the registered image. Image acquisition means, sharpness detection means for detecting the sharpness of the image, and the sharpness detection means detects the sharpness of the registered image and the input image, and the sharpness adjustment means detects at least one of the images. Sharpness difference reducing means for adjusting the sharpness to reduce the difference in sharpness between the registered image and the input image, and the registered image in which the difference in sharpness is reduced by the sharpness difference reducing means and the The image collating apparatus has image collating means for collating with the input image.
　また、請求項２の発明では、画像照合装置であって、画像の鮮鋭度を検出する鮮鋭度検出手段と、前記鮮鋭度検出手段により検出された鮮鋭度を調整する鮮鋭度調整手段と、前記鮮鋭度検出手段により検出される鮮鋭度が所定値に近づくように前記鮮鋭度調整手段によって調整された画像を登録画像として記憶した画像情報記憶手段と、前記登録画像と照合する入力画像を取得する画像取得手段と、前記画像取得手段により取得された入力画像の鮮鋭度を前記所定値に近づけるように調整して、該入力画像と前記登録画像との鮮鋭度の差を低減する鮮鋭度差低減手段と、前記鮮鋭度差低減手段により鮮鋭度の差が低減された前記入力画像と前記登録画像とを照合する画像照合手段とを有する画像照合装置とした。 The invention of claim 2 is an image collating device, wherein a sharpness detecting means for detecting a sharpness of an image, a sharpness adjusting means for adjusting a sharpness detected by the sharpness detecting means, An image information storage unit that stores an image adjusted by the sharpness adjustment unit as a registered image so that the sharpness detected by the sharpness detection unit approaches a predetermined value, and an input image to be compared with the registered image is acquired. A sharpness difference reduction that reduces the sharpness difference between the input image and the registered image by adjusting the sharpness of the input image acquired by the image acquisition means and the image acquisition means to approach the predetermined value And an image collating device that collates the input image whose sharpness difference is reduced by the sharpness difference reducing unit and the registered image.
　また、請求項３の発明では、画像照合装置であって、画像の鮮鋭度を検出する鮮鋭度検出手段と、前記鮮鋭度検出手段により検出された鮮鋭度を調整する鮮鋭度調整手段と、前記鮮鋭度検出手段により検出される鮮鋭度が所定値に近づくように前記鮮鋭度調整手段によって調整された登録画像より特徴を抽出した登録特徴情報を記憶した画像情報記憶手段と、前記登録画像と照合する入力画像を取得する画像取得手段と、前記画像取得手段により取得された前記入力画像の鮮鋭度を前記所定値に近づけるように調整して、前記入力画像と前記登録画像との鮮鋭度の差を低減する鮮鋭度差低減手段と、前記鮮鋭度差低減手段により鮮鋭度が調整された前記入力画像の特徴を抽出した入力特徴情報と前記登録特徴情報とを照合することにより前記入力画像と前記登録画像とを照合する画像照合手段とを有する画像照合装置とした。 The invention of claim 3 is an image collating device, wherein a sharpness detecting means for detecting a sharpness of an image, a sharpness adjusting means for adjusting a sharpness detected by the sharpness detecting means, Image information storage means storing registered feature information obtained by extracting features from the registered image adjusted by the sharpness adjustment means so that the sharpness detected by the sharpness detection means approaches a predetermined value, and collation with the registered image An image acquisition means for acquiring the input image, and adjusting the sharpness of the input image acquired by the image acquisition means so as to approach the predetermined value, and the difference in the sharpness between the input image and the registered image Sharpness difference reducing means for reducing the image quality, and input feature information obtained by extracting features of the input image whose sharpness has been adjusted by the sharpness difference reduction means and the registered feature information And an image matching device and an image matching unit for matching between the input image and the registered image.
　また、請求項４の発明では、前記画像照合手段は、前記登録画像と前記入力画像とを照合して、前記登録画像と前記入力画像との類似度を算出する画像照合装置とした。 In the invention of claim 4, the image collating means collates the registered image with the input image, and calculates an similarity between the registered image and the input image.
　また、請求項５の発明では、前記画像情報記憶手段から、前記入力画像と照合する登録画像に対応した情報を選択する照合対象選択手段をさらに備え、前記画像照合手段は、前記照合対象選択手段により選択された情報を用いて、前記入力画像の撮像対象と前記登録画像の撮像対象とが同一であるか否かを判定する画像照合装置とした。 Further, the invention of claim 5 further includes a collation target selection unit that selects information corresponding to a registered image to be collated with the input image from the image information storage unit, and the image collation unit includes the collation target selection unit. By using the information selected by the above, an image collating apparatus that determines whether the imaging target of the input image is the same as the imaging target of the registered image is used.
　また、請求項６の発明では、前記鮮鋭度差低減手段は、前記入力画像と前記登録画像との鮮鋭度の差が所定の差よりも大きいときに、前記入力画像と前記登録画像との鮮鋭度の差を低減する画像照合装置とした。 Further, in the invention of claim 6, the sharpness difference reducing means is configured to sharpen the input image and the registered image when a difference in sharpness between the input image and the registered image is larger than a predetermined difference. An image collation apparatus that reduces the difference in degrees was used.
　また、請求項７の発明では、画像照合方法であって、第１の画像と第２の画像とのうち少なくとも一方の画像の鮮鋭度を調整して、前記第１の画像と第２の画像との鮮鋭度の差を低減するステップと、鮮鋭度の差を低減した前記第１の画像と第２の画像とを照合するステップとを有する画像照合方法とした。 According to a seventh aspect of the present invention, there is provided an image collation method, wherein the first image and the second image are adjusted by adjusting a sharpness of at least one of the first image and the second image. And a step of collating the first image with the sharpness difference reduced with the second image.
　本発明によれば、画像照合装置であって、登録画像を記憶した画像情報記憶手段と、前記登録画像と照合する入力画像を取得する画像取得手段と、画像の鮮鋭度を検出する鮮鋭度検出手段と、前記鮮鋭度検出手段により前記登録画像及び前記入力画像の鮮鋭度を検出し、前記鮮鋭度調整手段により、少なくとも一方の画像の鮮鋭度を調整して、前記登録画像と前記入力画像との鮮鋭度の差を低減する鮮鋭度差低減手段と、前記鮮鋭度差低減手段により鮮鋭度の差が低減された前記登録画像と前記入力画像とを照合する画像照合手段と有する画像照合装置としたため、登録画像の鮮鋭度と入力画像の鮮鋭度とが、そのままでは画像の照合精度に影響する程に異なっていた場合であっても、画像の照合精度の低下度合いを小さくすることができる画像照合装置を提供することができる。 According to the present invention, there is provided an image collation apparatus, an image information storage unit that stores a registered image, an image acquisition unit that acquires an input image to be collated with the registered image, and a sharpness detection that detects the sharpness of the image. And the sharpness detecting means detects the sharpness of the registered image and the input image, and the sharpness adjusting means adjusts the sharpness of at least one of the images, and the registered image and the input image Sharpness difference reducing means for reducing the difference in sharpness of the image, and an image matching apparatus having image matching means for matching the registered image and the input image in which the difference in sharpness is reduced by the sharpness difference reducing means, and Therefore, even if the sharpness of the registered image and the sharpness of the input image are different so as to affect the matching accuracy of the image as it is, the degree of decrease in the matching accuracy of the image can be reduced. It is possible to provide an image collating apparatus.
　また、本発明によれば、画像照合装置であって、画像の鮮鋭度を検出する鮮鋭度検出手段と、前記鮮鋭度検出手段により検出された鮮鋭度を調整する鮮鋭度調整手段と、前記鮮鋭度検出手段により検出される鮮鋭度が所定値に近づくように前記鮮鋭度調整手段によって調整された画像を登録画像として記憶した画像情報記憶手段と、前記登録画像と照合する入力画像を取得する画像取得手段と、前記画像取得手段により取得された入力画像の鮮鋭度を前記所定値に近づけるように調整して、該入力画像と前記登録画像との鮮鋭度の差を低減する鮮鋭度差低減手段と、前記鮮鋭度差低減手段により鮮鋭度の差が低減された前記入力画像と前記登録画像とを照合する画像照合手段とを有する画像照合装置としたため、登録画像の鮮鋭度と入力画像の鮮鋭度とが、そのままでは画像の照合精度に影響する程に異なっていた場合であっても、画像の照合精度の低下度合いを小さくすることができる画像照合装置を提供することができる。また、照合時に登録画像の鮮鋭度を調整する必要がないので画像の照合に要する処理時間を短縮することができる。 Further, according to the present invention, there is provided an image collating apparatus, wherein a sharpness detection unit that detects a sharpness of an image, a sharpness adjustment unit that adjusts a sharpness detected by the sharpness detection unit, and the sharpness An image information storing means for storing an image adjusted by the sharpness adjusting means so that the sharpness detected by the degree detecting means approaches a predetermined value, and an image for acquiring an input image to be collated with the registered image And a sharpness difference reduction unit that adjusts the sharpness of the input image acquired by the image acquisition unit so as to approach the predetermined value and reduces a difference in the sharpness of the input image and the registered image. And an image collating device that collates the input image whose sharpness difference has been reduced by the sharpness difference reducing unit with the registered image, so that the sharpness of the registered image and the input image And sharpness of, is intact even when different enough to affect the matching precision of the image, it is possible to provide an image collating apparatus capable of reducing the deterioration degree of the matching precision of the image. In addition, since it is not necessary to adjust the sharpness of the registered image at the time of collation, the processing time required for image collation can be shortened.
　また、本発明によれば、画像照合装置であって、画像の鮮鋭度を検出する鮮鋭度検出手段と、前記鮮鋭度検出手段により検出された鮮鋭度を調整する鮮鋭度調整手段と、前記鮮鋭度検出手段により検出される鮮鋭度が所定値に近づくように前記鮮鋭度調整手段によって調整された登録画像より特徴を抽出した特徴情報を記憶した画像情報記憶手段と、前記登録画像と照合する入力画像を取得する画像取得手段と、前記画像取得手段により取得された前記入力画像の鮮鋭度を前記所定値に近づけるように調整して、前記入力画像と前記登録画像との鮮鋭度の差を低減する鮮鋭度差低減手段と、前記鮮鋭度差低減手段により鮮鋭度が調整された前記入力画像の特徴を抽出した入力特徴情報と前記登録特徴情報とを照合することにより前記入力画像と前記登録画像とを照合する画像照合手段とを有する画像照合装置としたため、登録画像の鮮鋭度と入力画像の鮮鋭度とが、そのままでは画像の照合精度に影響する程に異なっていた場合であっても、画像の照合精度の低下度合いを小さくすることができる画像照合装置を提供することができる。また、照合時に登録画像の鮮鋭度を調整して登録特徴情報を抽出する必要がないので、画像の照合に要する処理時間を短縮することができる。 Further, according to the present invention, there is provided an image collating apparatus, wherein a sharpness detection unit that detects a sharpness of an image, a sharpness adjustment unit that adjusts a sharpness detected by the sharpness detection unit, and the sharpness Image information storage means for storing feature information obtained by extracting features from the registered image adjusted by the sharpness adjustment means so that the sharpness detected by the degree detection means approaches a predetermined value, and input for collating with the registered image An image acquisition unit that acquires an image, and a sharpness of the input image acquired by the image acquisition unit is adjusted to be close to the predetermined value to reduce a difference in the sharpness between the input image and the registered image A sharpness difference reducing unit that matches the input feature information obtained by extracting features of the input image whose sharpness has been adjusted by the sharpness difference reducing unit and the registered feature information. Since the image collation apparatus has an image collation unit that collates the registered image with the registered image, the sharpness of the registered image and the sharpness of the input image are different from each other so that the collation accuracy of the image is affected as it is. Even if it exists, the image collation apparatus which can make the fall degree of the collation precision of an image small can be provided. In addition, since it is not necessary to adjust the sharpness of the registered image and extract the registered feature information at the time of collation, the processing time required for image collation can be shortened.
　また、本発明によれば、前記画像照合手段は、前記登録画像と前記入力画像とを照合して、前記登録画像と前記入力画像との類似度を算出する画像照合装置としたため、登録画像の鮮鋭度と入力画像の鮮鋭度とが、そのままでは画像の照合精度に影響する程に異なっていた場合であっても、画像の照合精度の低下度合いを小さくすることができ、入力画像と登録画像との類似度を算出することができる。 Further, according to the present invention, the image collating unit collates the registered image with the input image and calculates an similarity between the registered image and the input image. Even if the sharpness and the sharpness of the input image are different so as to affect the matching accuracy of the image as it is, the reduction degree of the matching accuracy of the image can be reduced, and the input image and the registered image can be reduced. The similarity can be calculated.
　また、本発明によれば、前記画像情報記憶手段から、前記入力画像と照合する登録画像に対応した情報を選択する照合対象選択手段をさらに備え、前記画像照合手段は、前記照合対象選択手段により選択された情報を用いて、前記入力画像の撮像対象と前記登録画像の撮像対象とが同一であるか否かを判定する画像照合装置としたため、登録画像の鮮鋭度と入力画像の鮮鋭度とが、そのままでは画像の照合精度に影響する程に異なっていた場合であっても、画像の照合精度の低下度合いを小さくすることができ、入力画像と登録画像との撮像対象が同一であるか否かを判定することができる。 According to the present invention, there is further provided a collation target selection unit that selects information corresponding to a registered image to be collated with the input image from the image information storage unit, and the image collation unit is selected by the collation target selection unit. Since the selected information is used as an image collating device that determines whether the imaging target of the input image and the imaging target of the registered image are the same, the sharpness of the registered image and the sharpness of the input image However, even if the image matching accuracy is different as it is, the degree of decrease in the image matching accuracy can be reduced, and whether the input image and the registered image are the same imaging target. It can be determined whether or not.
　また、本発明によれば、前記鮮鋭度差低減手段は、前記入力画像と前記登録画像との鮮鋭度の差が所定の差よりも大きいときに、前記入力画像と前記登録画像との鮮鋭度の差を低減する画像照合装置としたため、登録画像と入力画像の鮮鋭度の差が小さいときには、鮮鋭度差低減手段は、鮮鋭度の差の低減を行わないので、画像の照合に要する処理時間を短縮することができる。 Further, according to the present invention, the sharpness difference reducing means is configured such that when the difference in sharpness between the input image and the registered image is larger than a predetermined difference, the sharpness between the input image and the registered image. Therefore, when the difference in sharpness between the registered image and the input image is small, the sharpness difference reducing means does not reduce the difference in sharpness, so that the processing time required for image matching is reduced. Can be shortened.
　また、本発明によれば、画像照合方法であって、第１の画像と第２の画像とのうち少なくとも一方の画像の鮮鋭度を調整して、前記第１の画像と第２の画像との鮮鋭度の差を低減するステップと、鮮鋭度の差を低減した前記第１の画像と第２の画像とを照合するステップとを有する画像照合方法としたため、２つの画像の鮮鋭度の差が、そのままでは画像の照合精度に影響する程に異なっている場合でも、画像の照合精度の低下度合いを小さくして画像照合を行うことができる。 According to the present invention, there is also provided an image collating method, wherein the first image and the second image are adjusted by adjusting a sharpness of at least one of the first image and the second image. Since the image collating method has the step of reducing the difference in sharpness between the first image and the step of collating the first image and the second image in which the difference in sharpness is reduced, the difference in sharpness between the two images However, even if it is different as it is to the extent that it affects the image collation accuracy, image collation can be performed with a reduced degree of decrease in the image collation accuracy.
図１は、本実施形態に係る画像照合装置の構成を示す機能ブロック図である。FIG. 1 is a functional block diagram showing the configuration of the image collating apparatus according to this embodiment. 図２は、本実施形態における第１の高域フィルタ及び第２の高域フィルタを示す説明図である。FIG. 2 is an explanatory diagram showing the first high-pass filter and the second high-pass filter in the present embodiment. 図３は、本実施形態に係る画像照合装置による画像登録時及び画像照合時における画像処理の手順を示す説明図である。FIG. 3 is an explanatory diagram showing a procedure of image processing at the time of image registration and image verification by the image verification apparatus according to the present embodiment. 図４は、画像登録時に本実施形態に係る画像照合装置で実行される処理を示すフローチャートである。FIG. 4 is a flowchart showing processing executed by the image matching apparatus according to the present embodiment at the time of image registration. 図５は、画像入力時に本実施形態に係る画像照合装置で実行される処理を示すフローチャートである。FIG. 5 is a flowchart showing processing executed by the image collating apparatus according to the present embodiment when inputting an image. 図６は、画像入力時に本実施形態に係る画像照合装置で実行される画像照合処理を示すフローチャートである。FIG. 6 is a flowchart showing an image matching process executed by the image matching apparatus according to the present embodiment when inputting an image. 図７は、変形例に係る画像照合装置の構成を示す機能ブロック図である。FIG. 7 is a functional block diagram illustrating a configuration of an image collating apparatus according to a modification.
符号の説明Explanation of symbols
１　　　画像照合装置
２　　　撮像部
３　　　情報処理部
４　　　モード選択部
５　　　表示部
６　　　鮮鋭度検出部
７　　　鮮鋭度調整部
８　　　特徴情報抽出部
９　　　照合部
１０　　画像情報記憶部
１１　　登録画像選択部 DESCRIPTION OF SYMBOLS 1 Image collation apparatus 2 Imaging part 3 Information processing part 4 Mode selection part 5 Display part 6 Sharpness detection part 7 Sharpness adjustment part 8 Feature information extraction part 9 Collation part 10 Image information storage part 11 Registered image selection part
　以下、添付図面を参照して、本発明に係る画像照合装置及び画像照合方法の好適な一実施形態について説明する。本実施形態では、登録画像として複数人の人の耳を撮像した登録画像を予め登録しておき、その後、ある人の耳を撮像した入力画像と予め登録しておいた登録画像との照合を行う画像照合装置を例に挙げて説明する。なお、本発明に係る画像照合装置が照合する画像の撮像対象は、人の耳に限定するものではなく、人体の他の部位や人体以外のものとすることができる。 Hereinafter, a preferred embodiment of an image collation apparatus and an image collation method according to the present invention will be described with reference to the accompanying drawings. In the present embodiment, a registered image obtained by imaging a plurality of people's ears is registered in advance as a registered image, and thereafter, an input image obtained by imaging a certain person's ear and a registered image registered in advance are checked. An image collating apparatus to be performed will be described as an example. In addition, the imaging target of the image collated by the image collating apparatus according to the present invention is not limited to the human ear, and may be other parts of the human body or other than the human body.
　まず、図１を参照して本実施形態に係る画像照合装置の構成等について説明する。図１は、本実施形態に係る画像照合装置の構成を示す機能ブロック図である。図１に示すように、本実施形態に係る画像照合装置１は、撮像対象を撮像する撮像部２と、撮像部２により撮像した画像を登録画像として記憶して登録する登録処理や、画像照合を行うために後に撮像された画像（以下、「入力画像」という。）と登録画像との照合を行う照合処理等を行う情報処理部３と、この情報処理部３に登録処理を実行させる登録モードと照合処理を実行させる照合モードとを切替えるために操作するモード選択部４と、モード選択部４により選択したモードや情報処理部３による入力画像と登録画像との照合結果等を示す各種画像を表示する表示部５とを備えている。 First, the configuration of the image collating apparatus according to the present embodiment will be described with reference to FIG. FIG. 1 is a functional block diagram showing the configuration of the image collating apparatus according to this embodiment. As shown in FIG. 1, the image matching device 1 according to the present embodiment includes an imaging unit 2 that captures an imaging target, a registration process that stores and registers an image captured by the imaging unit 2 as a registered image, and image verification. Information processing unit 3 that performs a collation process or the like for collating an image (hereinafter referred to as an “input image”) captured later to perform registration with a registered image, and registration that causes the information processing unit 3 to perform a registration process Mode selection unit 4 operated to switch between the mode and the collation mode for executing the collation processing, and various images indicating the mode selected by the mode selection unit 4 and the collation result of the input image and the registered image by the information processing unit 3 Is displayed.
　撮像部２は、ＣＣＤ（Charge Coupled Device）カメラにより構成しており、撮像対象である人の耳を撮像して得たアナログの画像信号をデジタルの画像情報に変換して、情報処理部３へ出力する。 The imaging unit 2 is configured by a CCD (Charge-Coupled Device) camera, converts an analog image signal obtained by imaging a human ear that is an imaging target into digital image information, and sends the digital image information to the information processing unit 3. Output.
　本実施形態では、この撮像部２が登録画像や登録画像と照合する入力画像を撮像して、各画像に関する画像情報を取得する画像取得手段として機能する。なお、本実施形態では、撮像部２によって撮像された画像を登録画像として登録するように構成しているが、これに限らず、登録用に撮像された耳の写真をスキャナ等で読取って取得した写真の画像データや、デジタルカメラ等により別の場所で撮像した耳の画像データを登録画像として本実施形態の画像照合装置に登録するように構成してもよい。 In the present embodiment, the imaging unit 2 functions as an image acquisition unit that captures a registered image and an input image to be compared with the registered image and acquires image information regarding each image. In the present embodiment, the image picked up by the image pickup unit 2 is registered as a registered image. However, the present invention is not limited to this, and a photograph of an ear picked up for registration is read and acquired. The image data of the photograph or the image data of the ear captured at another location by a digital camera or the like may be registered as a registered image in the image collating apparatus of the present embodiment.
　情報処理部３は、ＣＰＵ（Central Processing Unit）とＲＯＭ（Read-Only Memory)とＲＡＭ（Random Access Memory）とを備えたマイクロコンピュータにより構成している。ＲＯＭには、情報処理部３が上記登録処理や照合処理等を行う際にＣＰＵが実行する各種プログラムが記憶されている。また、ＲＡＭは、ＣＰＵがＲＯＭに記憶されている各種プログラムを実行する際に作業領域として機能する一時記憶部である。 The information processing unit 3 includes a microcomputer having a CPU (Central Processing Unit), a ROM (Read-Only Memory), and a RAM (Random Access Memory). The ROM stores various programs executed by the CPU when the information processing unit 3 performs the registration process, the collation process, and the like. The RAM is a temporary storage unit that functions as a work area when the CPU executes various programs stored in the ROM.
　そして、この情報処理部３は、ＣＰＵがＲＯＭに記憶されている各種プログラムを読み出して実行することにより、撮像部２等により撮像した画像の輪郭の際立ち度合い（以下、「鮮鋭度」という。）を検出する鮮鋭度検出部６、撮像部２等により撮像した画像の鮮鋭度を調整する鮮鋭度調整部７、鮮鋭度を調整した画像からその画像の特徴量を示す特徴情報を抽出する特徴情報抽出部８、登録画像と入力画像との照合を行う照合部８等として機能する。 The information processing unit 3 reads and executes various programs stored in the ROM by the CPU, whereby the degree of conspicuousness of the image captured by the imaging unit 2 or the like (hereinafter referred to as “sharpness”). A sharpness detection unit 6 for detecting image quality, a sharpness adjustment unit 7 for adjusting the sharpness of an image captured by the imaging unit 2, etc., and feature information for extracting feature information indicating the feature amount of the image from the image whose sharpness has been adjusted. The extraction unit 8 functions as a collation unit 8 that collates the registered image and the input image.
　また、情報処理部３は、複数の登録画像に関する画像情報を記憶する画像情報記憶部１０を備えている。この画像情報記憶部１０は、ＨＤＤ（Hard disk drive）により構成している。なお、画像情報記憶部１０はＨＤＤに限定するものではなく、大容量の記憶装置であれば、フラッシュメモリ等の不揮発性メモリを用いてもよい。本実施形態では、この画像情報記憶部１０が鮮鋭度を所定値に近づくように調整された登録画像の特徴を示す登録特徴情報を記憶した画像情報記憶手段として機能する。 Further, the information processing unit 3 includes an image information storage unit 10 that stores image information regarding a plurality of registered images. The image information storage unit 10 is configured by an HDD (Hard disk drive). Note that the image information storage unit 10 is not limited to the HDD, and a nonvolatile memory such as a flash memory may be used as long as the storage device has a large capacity. In the present embodiment, the image information storage unit 10 functions as an image information storage unit that stores registered feature information indicating features of a registered image adjusted so that the sharpness approaches a predetermined value.
　この情報処理部３は、モード選択部４が操作されて登録モードが選択されているときに、撮像部２からデジタルの画像情報が入力されると、その画像情報に対応する画像が登録画像であると判断する。 When the digital image information is input from the imaging unit 2 when the registration mode is selected by operating the mode selection unit 4, the information processing unit 3 displays an image corresponding to the image information as a registered image. Judge that there is.
　そして、情報処理部３は、まず、鮮鋭度検出部６として機能し、入力された画像情報に基づいて、登録画像の鮮鋭度を検出する。このように情報処理部３は、画像情報に基づいて、撮像した画像の鮮鋭度を検出する鮮鋭度検出手段として機能する。 Then, the information processing unit 3 first functions as the sharpness detection unit 6 and detects the sharpness of the registered image based on the input image information. In this way, the information processing unit 3 functions as a sharpness detection unit that detects the sharpness of a captured image based on the image information.
　その後、情報処理部３は、鮮鋭度調整部７として機能して、検出した鮮鋭度を調整する。このとき、情報処理部３は、検出した鮮鋭度が後に詳述する所定値よりも高い場合、その鮮鋭度を所定値に近づけるように下げる処理を行う。すなわち、鮮明すぎる登録画像を若干ぼやかす処理を行うのである。一方、情報処理部３は、検出した鮮鋭度が所定値以下であった場合、鮮鋭度の調整は行わない。 After that, the information processing unit 3 functions as the sharpness adjusting unit 7 and adjusts the detected sharpness. At this time, when the detected sharpness is higher than a predetermined value, which will be described in detail later, the information processing unit 3 performs a process of reducing the sharpness so as to approach the predetermined value. That is, a process of slightly blurring a registered image that is too clear is performed. On the other hand, the information processing unit 3 does not adjust the sharpness when the detected sharpness is equal to or less than a predetermined value.
　ここで情報処理部３が行う鮮鋭度を低下させる処理は、撮像部２で撮像された画像を完全にぼやかす処理ではなく、順次登録する複数の登録画像をそれぞれ別の登録画像として後に判別できる程度に、各登録画像の鮮鋭度を下げる処理である。このように、情報処理部３は、鮮鋭度検出手段により検出された登録画像の鮮鋭度を所定値に近づくように調整する鮮鋭度調整手段として機能する。 Here, the process of reducing the sharpness performed by the information processing unit 3 is not a process of completely blurring the image picked up by the image pickup unit 2, but a plurality of registered images to be sequentially registered can be discriminated later as separate registered images. This is a process for reducing the sharpness of each registered image. As described above, the information processing unit 3 functions as a sharpness adjusting unit that adjusts the sharpness of the registered image detected by the sharpness detecting unit so as to approach a predetermined value.
　次に、情報処理部３は、特徴情報抽出部８として機能して、鮮鋭度を所定値まで下げた画像の画像情報に基づいて、その画像の特徴量を示す特徴情報を抽出し、その抽出した特徴情報を登録画像の画像情報（登録特徴情報）として画像情報記憶部１０に記憶させる。ここでいう特徴情報とは、例えば、画像中の特徴的な形状の位置や、画像の濃淡の変化度合いである。このように、情報処理部３は、鮮鋭度の差を低減した登録画像からその画像の特徴情報を抽出する特徴情報抽出手段として機能する。また、情報処理部３は、鮮鋭度の調整が不要であった登録画像についても、その画像の特徴情報を抽出して登録画像の画像情報として画像情報記憶部１０に記憶させる。 Next, the information processing unit 3 functions as the feature information extraction unit 8 to extract feature information indicating the feature amount of the image based on the image information of the image whose sharpness is lowered to a predetermined value, and the extraction The obtained feature information is stored in the image information storage unit 10 as image information of the registered image (registered feature information). The feature information here is, for example, the position of a characteristic shape in the image and the degree of change in the shading of the image. In this way, the information processing unit 3 functions as a feature information extraction unit that extracts feature information of an image from a registered image in which the difference in sharpness is reduced. The information processing unit 3 also extracts the feature information of the registered image that does not require sharpness adjustment and stores it in the image information storage unit 10 as image information of the registered image.
　そして、情報処理部３は、撮像部２から新たな画像情報が入力される度に、順次鮮鋭度検出部６、鮮鋭度調整部７、特徴情報抽出部８として機能して、画像情報記憶部１０に複数の登録画像の画像情報を登録する。 The information processing unit 3 sequentially functions as the sharpness detection unit 6, the sharpness adjustment unit 7, and the feature information extraction unit 8 each time new image information is input from the imaging unit 2, and the image information storage unit In 10, image information of a plurality of registered images is registered.
　なお、図１では、鮮鋭度検出部６検出した登録画像の鮮鋭度を鮮鋭度調整部７で所定値に近づくように調整した後、この鮮鋭度を調整された登録画像から該登録画像の特徴情報を抽出して画像情報記憶部１０に記憶させるように構成しているが、特徴情報ではなく鮮鋭度が所定値に近づくように調整された登録画像の画像情報を画像情報記憶部１０に記憶させるように構成してもよい。かかる構成とした場合、画像情報記憶部１０が、鮮鋭度調整手段によって鮮鋭度を所定値に近づくように調整された登録画像を記憶した画像情報記憶手段として機能することとなる。 In FIG. 1, the sharpness of the registered image detected by the sharpness detection unit 6 is adjusted so as to approach a predetermined value by the sharpness adjustment unit 7, and then the registered image is characterized from the registered image whose sharpness has been adjusted. Although the information is extracted and stored in the image information storage unit 10, not the feature information but the image information of the registered image adjusted so that the sharpness approaches a predetermined value is stored in the image information storage unit 10. You may comprise. In the case of such a configuration, the image information storage unit 10 functions as an image information storage unit that stores a registered image whose sharpness is adjusted to approach a predetermined value by the sharpness adjustment unit.
　一方、情報処理部３は、モード選択部４が操作されて照合モードが選択されているときに、撮像部２からデジタルの画像情報が入力されると、その画像情報に対応する画像が入力画像であると判断する。 On the other hand, when digital mode information is input from the imaging unit 2 when the mode selection unit 4 is operated and the collation mode is selected, the information processing unit 3 displays an image corresponding to the image information as an input image. It is judged that.
　そして、情報処理部３は、まず、鮮鋭度検出部６として機能して入力画像の鮮鋭度を検出し、その後、鮮鋭度調整部７として機能して、検出した鮮鋭度を調整する。このとき、情報処理部３は、上記登録モードのときと同様に、検出した鮮鋭度が後述の所定値よりも高い場合、その鮮鋭度を所定値に近づけるように下げる処理を行う。このように、本実施形態では、情報処理部３が、鮮鋭度検出手段により検出された入力画像の鮮鋭度を所定値に近づけるように調整して、該入力画像と登録画像との鮮鋭度の差を低減する鮮鋭度差低減手段として機能する。 The information processing unit 3 first functions as the sharpness detection unit 6 to detect the sharpness of the input image, and then functions as the sharpness adjustment unit 7 to adjust the detected sharpness. At this time, as in the registration mode, when the detected sharpness is higher than a predetermined value described later, the information processing unit 3 performs a process of reducing the sharpness so as to approach the predetermined value. As described above, in the present embodiment, the information processing unit 3 adjusts the sharpness of the input image detected by the sharpness detection unit so as to approach a predetermined value, and determines the sharpness of the input image and the registered image. It functions as a sharpness difference reducing means for reducing the difference.
　次に、情報処理部３は、上記登録モードのときと同様に、特徴情報抽出部８として機能して、鮮鋭度を調整した入力画像の特徴情報（入力特徴情報）を抽出する。なお、情報処理部３は、鮮鋭度の調整が不要であった場合、鮮鋭度を調整していない入力画像の画像情報に基づいて、その入力画像の特徴情報（入力特徴情報）を抽出する。このように、本実施形態では、情報処理部３が鮮鋭度の差を低減した入力画像からその画像の特徴情報を抽出する特徴情報抽出手段として機能する。 Next, the information processing unit 3 functions as the feature information extraction unit 8 to extract the feature information (input feature information) of the input image with the sharpness adjusted, as in the registration mode. In addition, when the sharpness adjustment is unnecessary, the information processing unit 3 extracts the feature information (input feature information) of the input image based on the image information of the input image whose sharpness is not adjusted. As described above, in this embodiment, the information processing unit 3 functions as a feature information extraction unit that extracts feature information of an image from an input image in which a difference in sharpness is reduced.
　その後、情報処理部３は、照合部９として機能して、入力画像から抽出した入力特徴情報と、画像情報記憶部１０に記憶している登録画像の登録特徴情報とを比較照合することによって、入力画像の撮像対象と登録画像の撮像対象とが同一のものであるか否かの判定を行う。本実施形態では、この情報処理部３が鮮鋭度差低減手段により鮮鋭度が調整された入力画像の特徴を示す入力特徴情報と登録特徴情報とを照合することにより、入力画像と登録画像とを照合する画像照合手段として機能する。 Thereafter, the information processing unit 3 functions as the matching unit 9 to compare and match the input feature information extracted from the input image with the registered feature information of the registered image stored in the image information storage unit 10. It is determined whether or not the imaging target of the input image and the imaging target of the registered image are the same. In the present embodiment, the information processing unit 3 compares the input feature information indicating the feature of the input image whose sharpness has been adjusted by the sharpness difference reducing unit with the registered feature information, thereby obtaining the input image and the registered image. It functions as image collating means for collation.
　このとき、情報処理部３は、画像情報記憶部１０に登録されている登録特徴情報を各登録画像単位で順次読み出し、その登録画像の登録特徴情報と入力画像の入力特徴情報とを順次比較照合する。そして、情報処理部３は、入力特徴情報と特徴量が一致する登録特徴情報に対応する登録画像が存在した場合に、その旨を示す画像を表示部に表示させる一方、一致する登録画像が存在しなかった場合には、その旨を示す画像を表示部に表示させる。 At this time, the information processing unit 3 sequentially reads the registered feature information registered in the image information storage unit 10 for each registered image, and sequentially compares the registered feature information of the registered image with the input feature information of the input image. To do. When there is a registered image corresponding to the registered feature information whose feature quantity matches the input feature information, the information processing unit 3 displays an image indicating that on the display unit, while there is a matching registered image. If not, an image indicating that is displayed on the display unit.
　このように、本実施形態の画像照合装置１では、登録画像を撮像した際に、その登録画像の鮮鋭度がばらついていた場合、情報処理部３が、各登録画像の鮮鋭度をほぼ所定値とするように一律化する調整を行うことによって、鮮鋭度のばらつきを低減し、鮮鋭度をほぼ一律とした各画像の特徴情報を登録画像の登録特徴情報として登録している。 As described above, in the image collation device 1 according to the present embodiment, when the registration image is captured and the sharpness of the registration image varies, the information processing unit 3 sets the sharpness of each registration image to a predetermined value. By making the adjustment to be uniform, the variation in sharpness is reduced, and the feature information of each image with the sharpness being almost uniform is registered as the registered feature information of the registered image.
　そして、この画像照合装置１では、入力画像を撮像した際に、情報処理部３が入力画像の鮮鋭度を登録画像の一律化された鮮鋭度と略等しくなるように調整し、この鮮鋭度を調整した後の入力画像の入力特徴情報と、登録画像の登録特徴情報とを比較して、入力画像と登録画像との照合を行うようにしている。 In the image collating apparatus 1, when the input image is captured, the information processing unit 3 adjusts the sharpness of the input image so that it is substantially equal to the uniform sharpness of the registered image. The input feature information of the input image after adjustment and the registered feature information of the registered image are compared, and the input image and the registered image are collated.
　すなわち、本実施形態では、情報処理部３が鮮鋭度検出手段により検出された登録画像及び入力画像の鮮鋭度の差のうち少なくともいずれか一方を調整して、登録画像と入力画像との鮮鋭度の差を低減する鮮鋭度差低減手段として機能するのである。 That is, in this embodiment, the information processing unit 3 adjusts at least one of the difference between the sharpness of the registered image and the input image detected by the sharpness detecting unit, and the sharpness between the registered image and the input image is adjusted. It functions as a sharpness difference reducing means for reducing the difference between the two.
　これにより、本実施形態の画像照合装置１は、撮像した登録画像の鮮鋭度と入力画像の鮮鋭度とが、そのままでは画像の照合精度に影響する程に異なっていた場合であっても、それらの鮮鋭度の差を低減して照合することができので、画像の照合精度が低下することがなく、同一の撮像対象を撮像した登録画像と入力画像であるにもかかわらず、それらの画像が異なる撮像対象を撮像したものであると誤認識することがない。 Thereby, even if the sharpness of the captured registered image and the sharpness of the input image are different so as to affect the matching accuracy of the image as they are, Since the difference in sharpness can be reduced and collation can be performed, the collation accuracy of the image is not lowered, and the images are registered even though they are the registered image and the input image obtained by imaging the same imaging target. There is no possibility of misrecognizing that a different imaging target is captured.
　ここで、情報処理部３が鮮鋭度検出部６として機能した際の鮮鋭度の検出方法について図２を参照して説明する。図２は、鮮鋭度の検出に用いる画像処理用のフィルタを示す説明図である。 Here, a method for detecting the sharpness when the information processing unit 3 functions as the sharpness detection unit 6 will be described with reference to FIG. FIG. 2 is an explanatory diagram showing an image processing filter used for detection of sharpness.
　情報処理部３は、鮮鋭度検出部６として機能する際、撮像部２から入力されるデジタルの画像情報に対して、図２に示す第１の高域フィルタと、第２の高域フィルタという周波数特性の異なる２種類の画像処理用のフィルタを作用させることによって、入力画像又は登録画像として撮像された画像の画像情報から、各画像の鮮鋭度を検出する。 When the information processing unit 3 functions as the sharpness detection unit 6, the digital image information input from the imaging unit 2 is referred to as a first high-pass filter and a second high-pass filter illustrated in FIG. 2. By applying two types of image processing filters having different frequency characteristics, the sharpness of each image is detected from image information of an image captured as an input image or a registered image.
　情報処理部３に入力される画像情報は、撮像した画像の各画素を複数の０又は１の二値情報で表したデジタルの情報である。情報処理部３は、撮像部２から入力される画像情報に第１の高域フィルタとしてのＦ１ＡとＦ１Ｂとを作用させ、その結果得られる情報と、撮像部２から入力される画像情報に第２の高域フィルタとしてのＦ２を作用させ、その結果得られる情報とを用いて、撮像した画像の鮮鋭度を検出する。 The image information input to the information processing unit 3 is digital information in which each pixel of a captured image is represented by a plurality of binary information of 0 or 1. The information processing unit 3 causes F1A and F1B as the first high-pass filter to act on the image information input from the imaging unit 2, and applies the first information to the information obtained as a result and the image information input from the imaging unit 2. F2 as a high-pass filter 2 is applied, and the sharpness of the captured image is detected using information obtained as a result.
　この第１の高域フィルタは、ソーベルの微分オペレータと呼ばれるものであり、画像のエッジを抽出するために広く用いられているオペレータである。そして、この第１の高域フィルタへ入力される画像情報を（入力）とし、第１の高域フィルタから出力される情報を（出力１）とした場合、（出力１）は、式：（出力１）＝|（入力）＊Ｆ１Ａ|＋|（入力）＊Ｆ１Ｂ|により算出される。ただし、この式において＊は畳み込みを意味しており、|Ｘ|は、Ｘの絶対値を意味している。 This first high-pass filter is called a Sobel differential operator, and is an operator that is widely used to extract image edges. When the image information input to the first high-pass filter is (input) and the information output from the first high-pass filter is (output 1), (output 1) is expressed by the formula: ( Output 1) = | (input) * F1A | + | (input) * F1B | However, in this expression, * means convolution, and | X | means the absolute value of X.
　また第２の高域フィルタは、ラプラシアンオペレータと呼ばれるものであり、これも、画像のエッジを抽出するために広く用いられているオペレータである。そして、この第２の高域フィルタへ入力される画像情報を（入力）とし、第２の高域フィルタから出力される情報を（出力２）とした場合、（出力２）は、式：（出力２）＝|（入力）＊Ｆ２|により算出される。この式においても、＊は畳み込みを意味しており、|Ｘ|は、Ｘの絶対値を意味している。そして、情報処理部３が検出する鮮鋭度を（鮮鋭度）とした場合、（鮮鋭度）は、式：（鮮鋭度）＝２×Σ（出力２）／Σ（出力１）により算出される。 Also, the second high-pass filter is called a Laplacian operator, which is also an operator that is widely used for extracting image edges. When the image information input to the second high-pass filter is (input) and the information output from the second high-pass filter is (output 2), (output 2) is expressed by the formula: ( Output 2) = | (input) * F2 | Also in this formula, * means convolution, and | X | means the absolute value of X. When the sharpness detected by the information processing unit 3 is (sharpness), (sharpness) is calculated by the formula: (sharpness) = 2 × Σ (output 2) / Σ (output 1). .
　次に、情報処理部３が鮮鋭度調整部７として機能した際の鮮鋭度の調整方法について説明する。鮮鋭度調整部７として機能する際、情報処理部３は、画像情報を平滑化する処理を行うことによって、鮮鋭度が所定値よりも高い画像の鮮鋭度を所定値とする。 Next, a method for adjusting the sharpness when the information processing unit 3 functions as the sharpness adjusting unit 7 will be described. When functioning as the sharpness adjusting unit 7, the information processing unit 3 performs a process of smoothing the image information, thereby setting the sharpness of an image whose sharpness is higher than a predetermined value as a predetermined value.
　すなわち、情報処理部３は、まず、撮像部２から入力された画像情報をフーリエ変換して、その結果得られた周波数Ｗの情報に乗数Ｍ（Ｗ）を掛ける。このとき、先に検出した入力画像又は登録画像の鮮鋭度をＳ、鮮鋭度の目標値（所定値）をＳ０とした場合、２次元周波数を（Ｕ、Ｖ）とすると、Ｗは、式Ｗ＝ＳＱＲＴ（Ｕ×Ｕ＋Ｖ×Ｖ）で表され、乗数Ｍ（Ｗ）は、式Ｍ（Ｗ）＝１－α×（Ｓ－Ｓ０）×Ｗで表される。ただし、αは制御パラメータである。その後、周波数Ｗの情報に乗数Ｍ（Ｗ）を掛けて得られた情報を逆フーリエ変換することによって、画像情報の平滑化を行う。 That is, the information processing unit 3 first performs Fourier transform on the image information input from the imaging unit 2 and multiplies the information on the frequency W obtained as a result by a multiplier M (W). At this time, when the sharpness of the previously detected input image or registered image is S and the target value (predetermined value) of the sharpness is S0, and the two-dimensional frequency is (U, V), W is expressed by the equation W = SQRT (U × U + V × V), and the multiplier M (W) is expressed by the equation M (W) = 1−α × (S−S0) × W. Where α is a control parameter. Thereafter, the information obtained by multiplying the information of the frequency W by the multiplier M (W) is subjected to inverse Fourier transform to smooth the image information.
　こうして情報処理部３は、撮像部２から入力された画像情報から所定量の高調波成分を除去することによって、画像情報を平滑化して、鮮鋭度が所定値よりも高い画像の鮮鋭度を所定値とする。 In this way, the information processing unit 3 smoothes the image information by removing a predetermined amount of harmonic components from the image information input from the imaging unit 2, and sets the sharpness of an image whose sharpness is higher than a predetermined value to a predetermined value. Value.
　次に、情報処理部３が特徴情報抽出部８として機能した際の特徴情報の抽出方法について説明する。特徴情報抽出部８として機能する際、情報処理部３は、まず、鮮鋭度を調整した画像情報に対応する画像をＭ×Ｎのブロックに分割する。 Next, a feature information extraction method when the information processing unit 3 functions as the feature information extraction unit 8 will be described. When functioning as the feature information extraction unit 8, the information processing unit 3 first divides the image corresponding to the image information whose sharpness has been adjusted into M × N blocks.
　そして、情報処理部３は、Ｍ×Ｎのブロックに分割した各ブロックに対応する情報に対して、順次Ｆ個のガボールフィルタを作用させ、各ブロックにＰ×Ｑ個のサンプル点を設定し、そのサンプル点におけるＦ個のデータを抽出する。ここで用いるガボールフィルタは、画像の輪郭線だけを抽出するために広く用いられているフィルタであり、ガボール関数と向きの合った輪郭線だけを抽出するフィルタである。こうして抽出したＦ個のデータを用いて、Ｐ×Ｑ×Ｆ次元の特徴量を取得し、この特徴量をその画像の特徴情報として抽出する。 Then, the information processing unit 3 sequentially applies F Gabor filters to information corresponding to each block divided into M × N blocks, sets P × Q sample points in each block, F data at the sample point are extracted. The Gabor filter used here is a filter that is widely used for extracting only the contour line of an image, and is a filter that extracts only a contour line whose direction matches the Gabor function. Using the F pieces of data thus extracted, a P × Q × F-dimensional feature amount is acquired, and this feature amount is extracted as feature information of the image.
　情報処理部３は、こうして抽出した登録画像の特徴情報（登録特徴情報）を各登録画像毎に、それぞれ画像情報記憶部１０に記憶させて登録しておき、入力画像が撮像されて、その画像に対応する画像情報が入力された場合に、その入力画像の特徴情報（入力特徴情報）と画像情報記憶部１０に登録されている登録画像の登録特徴情報とを比較して、入力画像と登録画像とを照合する。 The information processing unit 3 stores and registers the feature information (registered feature information) of the registered image extracted in this manner in the image information storage unit 10 for each registered image. When the image information corresponding to the input image is input, the feature information (input feature information) of the input image is compared with the registered feature information of the registered image registered in the image information storage unit 10, and the input image and the registered image are registered. Match the image.
　次に、上記のように構成した画像照合装置１による画像登録時及び画像照合時における画像処理の手順について、撮像画像を参照しながら説明する。図３は、画像登録時及び画像照合時における画像照合装置の画像処理手順を示す説明図である。なお、ここでは、一律化する鮮鋭度の所定値を０．２１として説明する。ここでの鮮鋭度の数値は、画像が鮮明であればあるほど、その値が大きくなる数値である。 Next, the procedure of image processing at the time of image registration and image collation by the image collating apparatus 1 configured as described above will be described with reference to captured images. FIG. 3 is an explanatory diagram showing an image processing procedure of the image collation apparatus at the time of image registration and image collation. Here, the description will be given assuming that the predetermined sharpness predetermined value is 0.21. The numerical value of the sharpness here is a numerical value that increases as the image becomes clearer.
　本実施形態の画像照合装置１では、予め複数の登録画像を記憶して登録しておき、その後に、入力画像を入力して入力画像と登録画像との照合を行う。まず、画像照合装置１が画像登録時に撮像した画像に対して行う画像処理の手順について説明する。 In the image collation apparatus 1 of the present embodiment, a plurality of registered images are stored and registered in advance, and thereafter, the input image is input and the input image and the registered image are collated. First, an image processing procedure performed on an image captured by the image matching device 1 at the time of image registration will be described.
　画像登録時に画像表示装置１は、図３に示すように、撮像対象である人の耳の画像を撮像する。その後、画像照合装置１では、撮像した画像の画像情報に基づいて、この画像の鮮鋭度を検出する。 At the time of image registration, the image display apparatus 1 captures an image of a person's ear that is an imaging target, as shown in FIG. Thereafter, the image matching device 1 detects the sharpness of the image based on the image information of the captured image.
　このとき、画像照合装置１は、検出した鮮鋭度が０．３１であった場合、検出した鮮鋭度が所定値である０．２１よりも高いため、その画像の鮮鋭度を０．２１まで下げる鮮鋭度調整処理を行う。一方、画像照合装置１は、撮像した画像の鮮鋭度が０．２１以下であった場合、鮮鋭度の調整を行わない。 At this time, if the detected sharpness is 0.31, the image matching apparatus 1 lowers the sharpness of the image to 0.21 because the detected sharpness is higher than a predetermined value of 0.21. Perform sharpness adjustment processing. On the other hand, the image matching device 1 does not adjust the sharpness when the sharpness of the captured image is 0.21 or less.
　そして、画像照合装置１は、鮮鋭度の調整を行った場合、鮮鋭度を調整した後の画像情報から、その画像の特徴情報を抽出する処理を行う。また、画像照合装置１は、撮像した画像に対して鮮鋭度の調整が不要であった場合、鮮鋭度の調整を行っていない画像情報から、その画像の特徴情報を抽出する処理を行う。 Then, when the sharpness adjustment is performed, the image matching device 1 performs a process of extracting feature information of the image from the image information after the sharpness adjustment. In addition, when the sharpness adjustment is not necessary for the captured image, the image matching device 1 performs a process of extracting feature information of the image from the image information that has not been subjected to the sharpness adjustment.
　画像照合装置１は、画像の登録時に撮像した複数の各画像に対して、上記のような画像処理を行った後、各画像の特徴情報を登録画像の登録特徴情報として画像情報記憶部１０に記憶させる。 The image matching device 1 performs the above-described image processing on each of the plurality of images captured at the time of image registration, and then stores the feature information of each image in the image information storage unit 10 as registered feature information of the registered image. Remember.
　次に、画像照合装置１が画像入力時に撮像した画像に対して行う画像処理の手順について説明する。画像入力時に画像表示装置１は、図３に示すように、撮像対象である人の耳の画像を撮像する。その後、画像照合装置１では、撮像した画像の画像情報に基づいて、この画像の鮮鋭度を検出する。 Next, a description will be given of the procedure of image processing performed on the image captured by the image matching device 1 when inputting an image. As shown in FIG. 3, the image display device 1 captures an image of a human ear that is an imaging target when inputting an image. Thereafter, the image matching device 1 detects the sharpness of the image based on the image information of the captured image.
　このとき、画像照合装置１は、検出した鮮鋭度が０．１９であった場合、検出した鮮鋭度が所定値である０．２１以下であるため、撮像した画像の鮮鋭度を調整する処理を行わない。一方、画像照合装置１は、このとき撮像した画像の鮮鋭度が所定値である０．２１よりも高かった場合、その画像の鮮鋭度を所定値である０．２１まで下げる処理を行う。 At this time, when the detected sharpness is 0.19, the image matching device 1 performs processing for adjusting the sharpness of the captured image because the detected sharpness is equal to or less than a predetermined value of 0.21. Not performed. On the other hand, when the sharpness of the image captured at this time is higher than a predetermined value of 0.21, the image collating apparatus 1 performs a process of reducing the sharpness of the image to a predetermined value of 0.21.
　その後、画像照合装置１は、鮮鋭度の調整が不要であった場合、鮮鋭度の調整を行っていない画像情報から、その画像の特徴情報を抽出する処理を行う。また、画像照合装置１は、鮮鋭度の調整を行った場合、鮮鋭度を調整した後の画像情報から、その画像の特徴情報を抽出する処理を行う。 Thereafter, when the sharpness adjustment is unnecessary, the image matching device 1 performs a process of extracting feature information of the image from the image information on which the sharpness is not adjusted. Further, when the sharpness is adjusted, the image matching device 1 performs a process of extracting feature information of the image from the image information after the sharpness is adjusted.
　そして、画像照合装置１は、こうして画像の入力時に撮像した画像の特徴情報を抽出すると、その画像の特徴情報である入力特徴情報と、既に画像情報記憶部１０に記憶している複数の登録特徴情報とを順次照合する処理を行い、その照合結果を表示部５へ出力する。 When the image collation apparatus 1 extracts the feature information of the image captured at the time of inputting the image in this way, the input feature information that is the feature information of the image and a plurality of registered features that are already stored in the image information storage unit 10. Processing for sequentially collating information is performed, and the collation result is output to the display unit 5.
　次に、上記画像登録時及び画像入力時に画像照合装置１で実行される処理について、図４～図６を参照して説明する。図４は、画像登録時に画像照合装置１で実行される処理を示すフローチャートであり、図５は、画像入力時に画像照合装置１で実行される処理を示すフローチャートであり、図６は、画像入力時に画像照合装置１で実行される画像照合処理を示すフローチャートである。 Next, processing executed by the image collating apparatus 1 at the time of image registration and image input will be described with reference to FIGS. FIG. 4 is a flowchart showing processing executed by the image matching device 1 at the time of image registration, FIG. 5 is a flowchart showing processing executed by the image matching device 1 at the time of image input, and FIG. 4 is a flowchart illustrating an image matching process that is sometimes executed by the image matching apparatus 1.
　まず、画像登録時に画像照合装置１で実行される処理について説明する。画像登録時に画像照合装置１では、撮像部２が撮像対象である人の耳を検知したか否かの判断を行い（ステップＳ１０１）、撮像対象を検知したと判断した場合（ステップＳ１０１：Ｙｅｓ）、その撮像対象を撮像して、撮像した登録画像に関する画像情報を情報処理部３へ出力する処理を行い、その後、処理をステップＳ１０２へ移す。一方、撮像部２が撮像対象を検知しなかった場合（ステップＳ１０１：Ｎｏ）、処理を終了する。 First, the process executed by the image collation apparatus 1 at the time of image registration will be described. At the time of image registration, the image collation apparatus 1 determines whether or not the imaging unit 2 has detected an ear of a person who is an imaging target (step S101) and determines that an imaging target has been detected (step S101: Yes). Then, the image capturing target is imaged, and image information related to the captured registered image is output to the information processing unit 3, and then the process proceeds to step S102. On the other hand, when the imaging unit 2 does not detect the imaging target (step S101: No), the process ends.
　ステップＳ１０２において、画像照合装置１では、情報処理部３のＣＰＵがＲＯＭから所定のプログラムを読み出して実行することにより、撮像部２から入力される登録画像の画像情報に基づいて、登録画像の鮮鋭度を検出する処理を行い、その後、処理をステップＳ１０３へ移す。 In step S <b> 102, in the image collation device 1, the CPU of the information processing unit 3 reads a predetermined program from the ROM and executes it, thereby sharpening the registered image based on the image information of the registered image input from the imaging unit 2. A process for detecting the degree is performed, and then the process proceeds to step S103.
　ステップＳ１０３において、画像照合装置１では、情報処理部３がステップＳ１０２で検出した登録画像の鮮鋭度が所定値である０．２１よりも高いか否かの判断を行い、０．２１よりも高いと判断した場合（ステップＳ１０３：Ｙｅｓ）、処理をステップＳ１０４へ移し、０．２１以下であると判断した場合（ステップＳ１０３：Ｎｏ）、処理をステップＳ１０５へ移す。 In step S103, the image collation apparatus 1 determines whether the sharpness of the registered image detected by the information processing unit 3 in step S102 is higher than a predetermined value of 0.21, and is higher than 0.21. If it is determined (step S103: Yes), the process proceeds to step S104. If it is determined that the value is 0.21 or less (step S103: No), the process proceeds to step S105.
　ステップＳ１０４において、画像照合装置１では、情報処理部３がステップＳ１０２で検出した鮮鋭度を所定値である０．２１まで下げる処理を行い、その後、処理をステップＳ１０５へ移す。このステップＳ１０５において、情報処理部３は、ステップＳ１０４で鮮鋭度の調整を行っていた場合、鮮鋭度の調整を行った後の画像情報からその登録画像の特徴情報（登録特徴情報）を抽出する処理を行い、鮮鋭度の調整を行わなかった場合には、撮像部２から入力された画像情報から登録画像の特徴情報（登録特徴情報）を抽出する処理を行う。その後、画像照合装置１では、情報処理部３がステップＳ１０５で抽出した登録特徴情報を登録画像の特徴量として画像情報記憶部１０に記憶させる処理を行って（ステップＳ１０６）、処理を終了する。 In step S104, in the image collation device 1, the information processing unit 3 performs a process of reducing the sharpness detected in step S102 to a predetermined value of 0.21, and then the process proceeds to step S105. In step S105, if the sharpness is adjusted in step S104, the information processing section 3 extracts the feature information (registered feature information) of the registered image from the image information after the sharpness adjustment. When the process is performed and the sharpness is not adjusted, the process extracts the registered image feature information (registered feature information) from the image information input from the imaging unit 2. Thereafter, in the image collation apparatus 1, the information processing unit 3 performs processing for storing the registered feature information extracted in step S105 in the image information storage unit 10 as the feature amount of the registered image (step S106), and ends the processing.
　画像照合装置１では、モード選択部４によって登録モードが選択されている間、撮像部２が撮像対象を検知する度に、上記ステップＳ１０１～ステップＳ１０６の処理が順次実行されて、画像情報記憶部１０に複数の登録画像の画像情報が登録される。なお、ここでは、画像情報記憶部１０に複数の登録画像の登録特徴情報を記憶させるようにしているが、登録特徴情報ではなく、鮮鋭度を所定値（ここでは、０．２１）に近づけるように調整した登録画像の画像情報を記憶させてもよい。 In the image collating apparatus 1, while the registration mode is selected by the mode selection unit 4, every time the imaging unit 2 detects an imaging target, the processing of step S101 to step S106 is sequentially executed, and the image information storage unit In FIG. 10, image information of a plurality of registered images is registered. Here, the registered feature information of a plurality of registered images is stored in the image information storage unit 10, but not the registered feature information but the sharpness is brought closer to a predetermined value (here, 0.21). The image information of the registered image that has been adjusted may be stored.
　次に、画像入力時に画像照合装置１で実行される処理について説明する。画像入力時に画像照合装置１では、撮像部２が撮像対象である人の耳を検知したか否かの判断を行い（ステップＳ２０１）、撮像対象を検知したと判断した場合（ステップＳ２０１：Ｙｅｓ）、その撮像対象を撮像して、撮像した入力画像に関する画像情報を情報処理部３へ出力する処理を行い、その後、処理をステップＳ２０２へ移す。一方、撮像部２が撮像対象を検知しなかった場合（ステップＳ２０１：Ｎｏ）、処理を終了する。 Next, processing executed by the image collation apparatus 1 when inputting an image will be described. At the time of image input, the image collation apparatus 1 determines whether or not the imaging unit 2 has detected an ear of a person who is an imaging target (step S201), and determines that an imaging target has been detected (step S201: Yes). Then, the image capturing target is imaged, and image information related to the captured input image is output to the information processing unit 3, and then the process proceeds to step S202. On the other hand, when the imaging unit 2 does not detect the imaging target (step S201: No), the process ends.
　ステップＳ２０２において、画像照合装置１では、情報処理部３のＣＰＵがＲＯＭから所定のプログラムを読み出して実行することにより、撮像部２から入力される入力画像の画像情報に基づいて、入力画像の鮮鋭度を検出する処理を行い、その後、処理をステップＳ２０３へ移す。 In step S202, in the image collation device 1, the CPU of the information processing unit 3 reads a predetermined program from the ROM and executes it, thereby sharpening the input image based on the image information of the input image input from the imaging unit 2. A process for detecting the degree is performed, and then the process proceeds to step S203.
　ステップＳ２０３において、画像照合装置１では、情報処理部３がステップＳ２０２で検出した入力画像の鮮鋭度が所定値である０．２１よりも高いか否かの判断を行い、０．２１よりも高いと判断した場合（ステップＳ２０３：Ｙｅｓ）、処理をステップＳ２０４へ移し、０．２１以下であると判断した場合（ステップＳ２０３：Ｎｏ）、処理をステップＳ２０５へ移す。 In step S203, the image matching device 1 determines whether or not the sharpness of the input image detected by the information processing unit 3 in step S202 is higher than a predetermined value of 0.21, and is higher than 0.21. If it is determined (step S203: Yes), the process proceeds to step S204. If it is determined that the value is 0.21 or less (step S203: No), the process proceeds to step S205.
　ステップＳ２０４において、画像照合装置１では、情報処理部３がステップＳ２０２で検出した鮮鋭度を所定値である０．２１まで下げる処理を行い、その後、処理をステップＳ２０５へ移す。このステップＳ２０５において、情報処理部３は、ステップＳ２０４で鮮鋭度の調整を行っていた場合、鮮鋭度の調整を行った後の画像情報からその入力画像の特徴情報（入力特徴情報）を抽出する処理を行い、鮮鋭度の調整を行わなかった場合には、撮像部２から入力された画像情報から入力画像の特徴情報（入力特徴情報）を抽出する処理を行い、その後、処理をステップＳ２０６へ移す。 In step S204, in the image collation device 1, the information processing unit 3 performs a process of lowering the sharpness detected in step S202 to a predetermined value of 0.21, and then the process proceeds to step S205. In step S205, if the sharpness adjustment is performed in step S204, the information processing unit 3 extracts the feature information (input feature information) of the input image from the image information after the sharpness adjustment. When the processing is performed and the sharpness is not adjusted, the processing is performed to extract the feature information (input feature information) of the input image from the image information input from the imaging unit 2, and then the processing proceeds to step S206. Transfer.
　ステップＳ２０６において、画像照合装置１では、情報処理部３が入力画像と登録画像とを比較照合する画像照合処理を行い、その後、処理を終了する。この画像照合処理については、次に図６を参照して具体的に説明する。 In step S206, in the image collation apparatus 1, the information processing unit 3 performs image collation processing for comparing and collating the input image and the registered image, and then ends the processing. The image collating process will be specifically described with reference to FIG.
　図６に示すように、画像照合処理が開始されると、画像照合装置１では、まず、情報処理部３がＲＡＭ内の所定量領域に変数Ｚとして「１」を設定する処理を行い（ステップＳ３０１）、その後、処理をステップＳ３０２へ移す。 As shown in FIG. 6, when the image matching process is started, in the image matching apparatus 1, first, the information processing unit 3 performs a process of setting “1” as the variable Z in a predetermined amount area in the RAM (step Thereafter, the process proceeds to step S302.
　ここで情報処理部３により設定される変数Ｚは、各登録画像が画像情報記憶部１０に登録されたときに情報処理部３によって順次割り当てられた識別番号であり、各登録画像を区別するための識別情報となるものである。そして、情報処理部３は、画像情報記憶部１０から登録画像の登録特徴情報を読み出す際に、Ｚ番目の登録画像の登録特徴情報を読み出す。 Here, the variable Z set by the information processing unit 3 is an identification number sequentially assigned by the information processing unit 3 when each registered image is registered in the image information storage unit 10, and is used for distinguishing each registered image. This is the identification information. When the information processing unit 3 reads the registered feature information of the registered image from the image information storage unit 10, the information processing unit 3 reads the registered feature information of the Zth registered image.
　ステップＳ３０２において、画像照合装置１では、情報処理部３が画像情報記憶部１０に登録されたＺ番目の登録画像の登録特徴情報を読み出した後、入力画像の入力特徴情報と、登録画像の登録特徴情報とを比較照合する処理を行い（ステップＳ３０３）、その後、処理をステップＳ３０４へ移す。 In step S302, in the image matching device 1, after the information processing unit 3 reads the registered feature information of the Zth registered image registered in the image information storage unit 10, the input feature information of the input image and the registration of the registered image are registered. A process of comparing and collating with feature information is performed (step S303), and then the process proceeds to step S304.
　なお、ここでは、入力特徴情報と登録特徴情報を比較照合することによって、入力画像と登録画像とを照合しているが、上記のように、画像情報記憶部１０に登録特徴情報ではなく、鮮鋭度を所定値に近づけるように調整した後の登録画像の画像情報を記憶させておいた場合には、この画像情報記憶部１０に記憶させた鮮鋭度を調整した後の登録画像の画像情報と、同じく鮮鋭度を所定値に近づけるように調整した後の入力画像の画像情報とを比較照合して、入力画像と登録画像の照合を行う。 In this case, the input feature information and the registered feature information are compared and collated to collate the input image and the registered image. However, as described above, the image information storage unit 10 is not registered feature information but sharp. When the image information of the registered image after adjusting the degree to approach the predetermined value is stored, the image information of the registered image after adjusting the sharpness stored in the image information storage unit 10 and Similarly, the input image and the registered image are collated by comparing and collating with the image information of the input image after adjusting the sharpness to be close to a predetermined value.
　ステップＳ３０４において、画像照合装置１では、情報処理部３がステップＳ３０３で比較照合した入力画像の入力特徴情報と登録画像の登録特徴情報とが一致したか否かの判断を行い、一致したと判断した場合（ステップＳ３０４：Ｙｅｓ）、処理をステップＳ３０７へ移して、入力画像の入力特徴情報と登録画像の登録特徴情報とが一致したことを示す画像を表示部５に表示させる照合結果表示処理を行い、その後、処理を終了する。 In step S304, the image matching apparatus 1 determines whether or not the input feature information of the input image compared and matched by the information processing unit 3 in step S303 matches the registered feature information of the registered image. If so (step S304: Yes), the process proceeds to step S307, and a matching result display process for displaying an image indicating that the input feature information of the input image matches the registered feature information of the registered image on the display unit 5 is performed. After that, the process ends.
　一方、情報処理部３が、ステップＳ３０３で比較照合した入力画像の入力特徴情報と登録画像の登録特徴情報とが一致しなかったと判断した場合（ステップＳ３０４：Ｎｏ）、処理をステップＳ３０５へ移し、変数Ｚに「１」を加算する処理を行い、その後、処理をステップＳ３０６へ移す。 On the other hand, when the information processing unit 3 determines that the input feature information of the input image compared and collated in step S303 does not match the registered feature information of the registered image (step S304: No), the process proceeds to step S305. A process of adding “1” to the variable Z is performed, and then the process proceeds to step S306.
　ステップＳ３０６において、画像照合装置１では、情報処理部３がステップＳ３０５で変数Ｚに「１」を加算した結果、変数Ｚが最大値Ｚｍａｘ（登録されている登録画像の数）に達したか否かの判断を行い、Ｚｍａｘに達していないと判断した場合（ステップＳ３０６：Ｎｏ）、処理をステップＳ３０２へ移し、Ｚｍａｘに達したと判断した場合（ステップＳ３０６：Ｙｅｓ）、処理をステップＳ３０７へ移す。 In step S306, in the image collation device 1, whether or not the variable Z has reached the maximum value Zmax (the number of registered images registered) as a result of the information processing unit 3 adding “1” to the variable Z in step S305. If it is determined that Zmax has not been reached (step S306: No), the process proceeds to step S302. If it is determined that Zmax has been reached (step S306: Yes), the process proceeds to step S307. .
　そして、ステップＳ３０７において、画像照合装置１では、情報処理部３がステップＳ３０６において変数ＺがＺｍａｘに達したと判断されていた場合、入力画像の入力特徴情報と一致する登録特徴情報を持つ登録画像が存在しないことを示す画像を表示部５に表示させる照合結果表示処理を行い、その後、処理を終了する。 In step S307, in the image collation device 1, when the information processing unit 3 determines that the variable Z has reached Zmax in step S306, the registered image having the registered feature information that matches the input feature information of the input image. A collation result display process for displaying an image indicating that no is present on the display unit 5 is performed, and then the process ends.
　なお、本実施形態では、ステップＳ３０４で入力画像の入力特徴情報と登録画像の登録特徴情報とが一致したか否かを判断するように構成しているが、このステップＳ３０４での判断を行わずに、全ての登録画像と入力画像との画像情報同士を照合した後に、入力画像の入力特徴情報と登録画像の登録特徴情報が一致したか否かを判断するように構成してもよい。 In the present embodiment, it is configured to determine whether or not the input feature information of the input image matches the registered feature information of the registered image in step S304, but the determination in step S304 is not performed. In addition, after collating image information of all registered images and input images, it may be configured to determine whether or not the input feature information of the input image matches the registered feature information of the registered image.
　また、本実施形態では、ステップＳ３０７の照合結果表示処理において、入力画像と登録画像の照合結果、すなわち、入力画像と登録画像とが一致していたこと、若しくは、入力画像と一致する登録画像が存在しないことのみを表示部５に表示させているが、情報処理部３によって、入力画像と全ての登録画像とを順次照合して、入力画像と各登録画像との類似度を算出し、入力画像との類似度が高い順に複数の登録画像を並べて表示部５に表示させるように構成してもよい。また、かかる構成とした場合、入力画像との類似度が所定の閾値を超える登録画像のみを、類似度の高い順に表示部５に表示させるようにしてもよい。 In the present embodiment, in the matching result display processing in step S307, the matching result between the input image and the registered image, that is, the input image and the registered image match, or the registered image that matches the input image is displayed. Although only the non-existence is displayed on the display unit 5, the information processing unit 3 sequentially collates the input image with all the registered images, calculates the similarity between the input image and each registered image, and inputs A plurality of registered images may be arranged and displayed on the display unit 5 in descending order of similarity to images. Further, in such a configuration, only registered images whose similarity to the input image exceeds a predetermined threshold may be displayed on the display unit 5 in descending order of similarity.
　以上説明したように、本実施形態の画像照合装置１は、登録用に予め撮像した複数の画像のそれぞれについて各画像の鮮鋭度を検出し、鮮鋭度が所定値よりも高い場合に、その鮮鋭度を所定値となるまで下げることによって、撮像した各画像の鮮鋭度を略同一となるように一律化した後、その画像の登録特徴情報を登録画像の特徴量として画像情報記憶１０に予め記憶させて登録する。 As described above, the image collation apparatus 1 according to the present embodiment detects the sharpness of each image for each of a plurality of images captured in advance for registration, and when the sharpness is higher than a predetermined value, the sharpness is determined. By lowering the degree to a predetermined value, the sharpness of each captured image is made uniform so as to be substantially the same, and then the registered feature information of the image is stored in advance in the image information storage 10 as the feature amount of the registered image. Let me register.
　そして、画像照合装置１は、入力用に撮像した画像についても、その画像の鮮鋭度を検出し、鮮鋭度が所定値よりも高い場合に限りその鮮鋭度を所定値となるまで下げてから、その画像の入力特徴情報を入力画像の特徴量として抽出して、入力画像の入力特徴情報と登録画像の登録特徴情報とを比較して、入力画像と登録画像との照合を行うように構成している。 Then, the image matching device 1 detects the sharpness of the image captured for input, and reduces the sharpness to a predetermined value only when the sharpness is higher than the predetermined value. The input feature information of the image is extracted as the feature quantity of the input image, and the input feature information of the input image is compared with the registered feature information of the registered image, and the input image and the registered image are compared. ing.
　そのため、この画像照合装置１では、入力画像と登録画像とを照合する際に、各画像の鮮鋭度が略同一な状態となっているため、撮像時に登録画像の鮮鋭度と入力画像の鮮鋭度とが、画像の照合精度に影響を及ぼすほどに異なっていた場合であっても、画像の照合精度が低下することがない。 Therefore, in this image collation apparatus 1, when the input image and the registered image are collated, the sharpness of each image is substantially the same. Therefore, the sharpness of the registered image and the sharpness of the input image are captured at the time of imaging. However, even if it is different so as to affect the collation accuracy of the image, the collation accuracy of the image is not lowered.
　なお、上記した実施形態では、一つの入力画像と複数の登録画像とを順次照合する場合を例に挙げて説明したが、本実施形態の画像照合装置１によれば、画像情報記憶部１０に一つの登録画像に対応する画像情報を登録しておき、複数の入力画像を当該画像照合装置１へ入力して、その複数の入力画像の中から登録画像の撮像対象と同一の撮像対象が撮像された入力画像を判定することもできる。 In the above-described embodiment, the case where one input image and a plurality of registered images are sequentially verified has been described as an example. However, according to the image verification device 1 of the present embodiment, the image information storage unit 10 includes Image information corresponding to one registered image is registered, a plurality of input images are input to the image collating apparatus 1, and the same imaging target as that of the registered image is captured from the plurality of input images. It is also possible to determine the input image.
　また、上記した画像照合装置１では、入力画像を撮像した場合に、その入力画像の入力特徴情報と特徴情報が一致する登録画像が見つかるまで、入力画像と登録画像との照合を繰り返し行うように構成したが、入力画像を撮像する際に、予めその入力画像と照合する登録画像を選択して、その登録画像と入力画像とを照合するように画像照合装置を構成してもよい。 Further, in the image collating apparatus 1 described above, when an input image is captured, the collation between the input image and the registered image is repeatedly performed until a registered image that matches the input feature information and the feature information of the input image is found. Although configured, when an input image is captured, a registered image to be collated with the input image may be selected in advance, and the image collating apparatus may be configured to collate the registered image with the input image.
　ここで、入力画像と照合する登録画像を選択可能に構成した画像照合装置について、図７を参照して説明する。図７は、変形例に係る画像照合装置を示す機能ブロック図である。なお、図７では、図１に示した画像照合装置と同様の機能を有する部分に同一の符号を付している。 Here, an image collating apparatus configured to be able to select a registered image to be collated with an input image will be described with reference to FIG. FIG. 7 is a functional block diagram illustrating an image collating apparatus according to a modification. In FIG. 7, the same reference numerals are given to parts having the same functions as those of the image collating apparatus shown in FIG. 1.
　図７に示すように、本変形例に係る画像照合装置１ａは、その基本的な構成に関して、図１に示した画像照合装置１と同様であるが、登録画像選択部１１を備えているところが図１に示した画像照合装置１と異なる。 As shown in FIG. 7, the image collation apparatus 1a according to the present modification is similar to the image collation apparatus 1 shown in FIG. 1 with respect to its basic configuration, but includes a registered image selection unit 11. Different from the image collating apparatus 1 shown in FIG.
　すなわち、この変形例に係る画像照合装置１ａは、図１に示した画像照合装置１と同様に、撮像部２と、情報処理部３とモード選択部４と表示部５とを備えている。また、情報処理部３は、内部のＣＰＵがＲＯＭに記憶している所定のプログラムを実行することによって、鮮鋭度検出部６、鮮鋭度調整部７、特徴情報抽出部８、照合部９等として機能するものである。また、この情報処理部３は、複数の登録画像の画像情報を記憶する画像情報記憶部１０ａを備えている。 That is, the image collation apparatus 1a according to this modification includes an imaging unit 2, an information processing unit 3, a mode selection unit 4, and a display unit 5, similarly to the image collation device 1 shown in FIG. In addition, the information processing unit 3 executes a predetermined program stored in the ROM by the internal CPU, so that the sharpness detection unit 6, the sharpness adjustment unit 7, the feature information extraction unit 8, the collation unit 9, and the like It functions. The information processing unit 3 also includes an image information storage unit 10a that stores image information of a plurality of registered images.
　図１に示した画像照合装置１の画像情報記憶部１０では、複数の登録画像の各特徴情報だけを記憶させていたが、この変形例に係る画像照合装置１ａの画像情報記憶部１０ａには、登録用として撮像した画像、又は予め別の場所で撮像された登録用の画像に関する全ての情報を各画像毎に登録画像の画像情報として記憶させている。この画像情報記憶部１０に記憶されている画像情報は、鮮鋭度の調整が行われていない登録画像そのものの画像情報である。 In the image information storage unit 10 of the image collating apparatus 1 shown in FIG. 1, only the feature information of a plurality of registered images is stored, but the image information storing unit 10a of the image collating apparatus 1a according to this modification is stored. All the information related to the image captured for registration or the image for registration captured in advance at another location is stored as image information of the registered image for each image. The image information stored in the image information storage unit 10 is image information of the registered image itself that has not been subjected to sharpness adjustment.
　そして、この画像照合装置１ａは、入力画像を撮像するときに、画像情報記憶部１０ａに記憶している複数の登録画像の画像情報の中から、入力画像と照合する特定の登録画像の画像情報を選択するために操作する登録画像選択部１１を備えている。本実施形態では、この変形例における登録画像選択部１１が、画像情報記憶手段から入力画像と照合する登録画像に対応した情報を選択する照合対象選択手段として機能するものである。 And when this image collation apparatus 1a images an input image, it is the image information of the specific registration image collated with an input image from the image information of the some registration image memorize | stored in the image information storage part 10a. A registered image selection unit 11 operated to select the image. In this embodiment, the registered image selection unit 11 in this modification functions as a collation target selection unit that selects information corresponding to a registered image to be collated with the input image from the image information storage unit.
　この画像照合装置１ａでは、予め登録用の画像を撮像する際に、撮像した登録画像に対してそれぞれ固有の識別番号（たとえば、上記変数Ｚ）を割り当てると共に、その識別番号を表示部５に表示させる。 In this image collation apparatus 1a, when a registration image is captured in advance, a unique identification number (for example, the variable Z) is assigned to each captured registered image, and the identification number is displayed on the display unit 5. Let
　そして、入力画像を撮像する際には、登録画像選択部１１を操作して、登録画像の登録時に表示部５に表示されていた識別番号を選択して入力する。この識別番号が入力されると、情報処理部３は、入力画像の照合対象として、入力された識別番号に対応する登録画像の画像情報を画像情報記憶部１０ａから読み出し、この読み出した画像情報に基づいて、登録画像の鮮鋭度の検出、鮮鋭度の調整、特徴情報の抽出を行い、入力画像の特徴情報と選択された登録画像の特徴情報とを比較して入力画像と登録画像との照合を行って、入力画像の撮像対象と登録画像の撮像対象とが同一であるか否かの判定を行う。 Then, when an input image is captured, the registered image selection unit 11 is operated to select and input the identification number displayed on the display unit 5 when registering the registered image. When this identification number is input, the information processing unit 3 reads out the image information of the registered image corresponding to the input identification number as a target for collation of the input image from the image information storage unit 10a. Based on this, the sharpness of the registered image is detected, the sharpness is adjusted, and the feature information is extracted. The feature information of the input image is compared with the feature information of the selected registered image, and the input image and the registered image are compared. To determine whether or not the imaging target of the input image is the same as the imaging target of the registered image.
　このように、本変形例に係る画像照合装置１ａでは、入力画像と照合する登録画像を予め選択し、その登録画像の画像情報に対して、必要に応じて鮮鋭度の調整を行ってから特徴情報を抽出して、入力画像の特徴情報と比較するように構成している。 As described above, the image collating apparatus 1a according to the present modification is characterized by selecting a registered image to be collated with the input image in advance and adjusting the sharpness of the image information of the registered image as necessary. Information is extracted and compared with the feature information of the input image.
　そのため、この画像照合装置１ａによれば、入力画像と特徴情報が一致する登録画像が見つかるまで、何度も画像照合処理を行う必要がないので、画像照合処理に要する処理時間を短縮することができる。 Therefore, according to the image collating apparatus 1a, it is not necessary to perform the image collating process repeatedly until a registered image whose feature information matches the input image is found, so that the processing time required for the image collating process can be shortened. it can.
　しかも、この変形例に係る画像照合装置１ａでは、選択された登録画像の鮮鋭度を所定値となるように調整すると共に、入力画像に対しても、必要に応じて鮮鋭度を所定値となるように調整するので、入力画像と登録画像とを照合する際には、入力画像と登録画像との鮮鋭度が略同一な状態となり、撮像時に登録画像の鮮鋭度と入力画像の鮮鋭度とが、画像の照合精度に影響を及ぼす程に異なっていた場合であっても、画像の照合精度が低下することがない。 Moreover, in the image collating apparatus 1a according to this modification, the sharpness of the selected registered image is adjusted to a predetermined value, and the sharpness of the input image is also set to a predetermined value as necessary. Therefore, when the input image and the registered image are collated, the sharpness of the input image and the registered image is substantially the same, and the sharpness of the registered image and the sharpness of the input image are obtained at the time of imaging. Even if they differ to such an extent that they affect the collation accuracy of the image, the collation accuracy of the image does not decrease.
　また、この変形例に係る画像照合装置１ａの画像情報記憶部１０ａには、上記したように、登録用の画像を撮像した際、撮像部２から情報処理部３へ入力される登録画像に関する全ての画像情報が記憶されているため、入力画像と登録画像とを照合する際に、登録画像の鮮鋭度を所定値よりも高くするように鮮鋭度を調整することができる。 In addition, as described above, the image information storage unit 10a of the image collating apparatus 1a according to this modification includes all of the registered images input from the imaging unit 2 to the information processing unit 3 when a registration image is captured. Therefore, when collating the input image with the registered image, the sharpness of the registered image can be adjusted to be higher than a predetermined value.
　すなわち、画像情報記憶部１０ａには、図１に示す画像情報記憶部１０のように、鮮鋭度を一律に調整した画像の特徴情報を記憶させるのではなく、鮮鋭度を調整する前の画像情報を記憶させている。そのため、登録画像の鮮鋭度を比較的柔軟に変更することができる。 That is, the image information storage unit 10a does not store the feature information of the image with the sharpness adjusted uniformly as in the image information storage unit 10 shown in FIG. 1, but the image information before adjusting the sharpness. Is remembered. Therefore, the sharpness of the registered image can be changed relatively flexibly.
　具体的には、入力画像の鮮鋭度が所定値よりも高く、登録画像選択部１１により選択された登録画像の鮮鋭度が入力画像の鮮鋭度とほぼ同一であった場合、情報処理部３は、その登録画像の鮮鋭度を所定値まで下げることなく比較的高く保った状態で、入力画像の特徴情報と登録画像の特徴情報とを比較することができる。これにより、鮮鋭度が比較的高い入力画像と登録画像とを照合することができるので、画像照合の精度をさらに向上させることができる。また、入力画像の鮮鋭度が所定値よりも低い場合に、登録画像の鮮鋭度を入力画像の鮮鋭度に合わせることもできる。これによっても、照合する際に入力画像と登録画像の鮮鋭度同士の差を低減することができるので、画像照合の精度を向上させることができる。 Specifically, if the sharpness of the input image is higher than a predetermined value and the sharpness of the registered image selected by the registered image selection unit 11 is substantially the same as the sharpness of the input image, the information processing unit 3 The feature information of the input image can be compared with the feature information of the registered image in a state where the sharpness of the registered image is kept relatively high without lowering to a predetermined value. As a result, the input image having a relatively high sharpness and the registered image can be collated, so that the accuracy of image collation can be further improved. In addition, when the sharpness of the input image is lower than a predetermined value, the sharpness of the registered image can be matched with the sharpness of the input image. This can also reduce the difference between the sharpnesses of the input image and the registered image when collating, thereby improving the accuracy of image collation.
　また、入力画像の鮮鋭度が選択された登録画像の鮮鋭度よりも更に高かった場合、本変形例に係る画像照合装置１ａでは、選択された登録画像の画像情報を用いて、その登録画像のエッジ部分を強調する画像処理を行い、入力画像の鮮鋭度に近づけてから特徴情報を抽出し、その特徴情報と入力画像の特徴情報とを比較することもできる。これにより、画像照合の精度をさらに向上させることができる。 Further, when the sharpness of the input image is higher than the sharpness of the selected registered image, the image collating apparatus 1a according to the present modification uses the image information of the selected registered image to display the registered image. It is also possible to perform image processing that emphasizes the edge portion, extract feature information after approaching the sharpness of the input image, and compare the feature information with the feature information of the input image. Thereby, the precision of image collation can be further improved.
　また、この変形例の画像照合装置１ａ、図１に示した画像照合装置１のいずれにおいても、入力用の画像を撮像した際には、撮像部２が入力画像に関する全ての画像情報を情報処理部３へ入力するように構成しているため、入力画像の鮮鋭度が所定値よりも低かった場合に、撮像部２から入力される入力画像の画像情報を用いて、その入力画像のエッジ部分を強調する画像処理を行い、入力画像の鮮鋭度を所定値に近づけてから特徴情報を抽出し、その特徴情報と選択された登録画像の特徴情報とを比較することもできる。これによって、画像照合の精度をより一層向上させることができる。 Further, in both the image collating apparatus 1a of this modification and the image collating apparatus 1 shown in FIG. 1, when an image for input is captured, the imaging unit 2 processes all image information related to the input image. Since the image is input to the unit 3, when the sharpness of the input image is lower than a predetermined value, the edge portion of the input image is used by using the image information of the input image input from the imaging unit 2. It is also possible to perform image processing for emphasizing the image, extract the feature information after bringing the sharpness of the input image close to a predetermined value, and compare the feature information with the feature information of the selected registered image. Thereby, the accuracy of image collation can be further improved.
　上記実施形態では、入力画像の鮮鋭度と登録画像の鮮鋭度との差を低減する一例として、入力画像の鮮鋭度と登録画像の鮮鋭度とが所定値よりも高い場合に、その鮮鋭度を所定となるように調整しているが、本発明の鮮鋭度差低減手段による鮮鋭度差の低減方法はこれに限定するものではなく、入力画像の鮮鋭度と登録画像の鮮鋭度との差を低減する方法であれば、任意の方法であってもよい。 In the above embodiment, as an example of reducing the difference between the sharpness of the input image and the sharpness of the registered image, when the sharpness of the input image and the sharpness of the registered image are higher than a predetermined value, the sharpness is calculated. Although it is adjusted so as to be predetermined, the method of reducing the sharpness difference by the sharpness difference reducing means of the present invention is not limited to this, and the difference between the sharpness of the input image and the sharpness of the registered image is calculated. Any method may be used as long as the method is reduced.
　たとえば、登録画像及び入力画像の鮮鋭度を所定値に一致させるのではなく、所定値とほぼ同程度の鮮鋭度となるように調整してもよい。また、鮮鋭度に所定値を設定せずに、入力画像と登録画像の２画像のうち、鮮鋭度の低い画像の方に合わせて鮮鋭度の高い方の画像の鮮鋭度を低下させてもよく、２画像のうち鮮鋭度の高い方の画像に合わせて鮮鋭度の低い方の画像の鮮鋭度を高めてもよい。 For example, instead of matching the sharpness of the registered image and the input image with a predetermined value, the sharpness may be adjusted to be substantially the same as the predetermined value. Further, without setting a predetermined value for the sharpness, the sharpness of the image with the higher sharpness may be reduced in accordance with the image with the lower sharpness of the input image and the registered image. The sharpness of the lower sharpness image may be increased in accordance with the higher sharpness image of the two images.
　さらには、入力画像の鮮鋭度と登録画像の鮮鋭度との差が所定の差よりも大きいときに、入力画像と登録画像との鮮鋭度の差を低減するように鮮鋭度差低減手段を構成してもよい。 Furthermore, the sharpness difference reducing means is configured to reduce the difference in sharpness between the input image and the registered image when the difference between the sharpness of the input image and the registered image is larger than a predetermined difference. May be.
　また、上記実施形態では、撮像部２で撮像した画像の画像情報を登録画像として画像情報記憶部１０に記憶して登録するようにしているが、画像情報記憶部１０に登録する登録画像は、必ずしも撮像部２で撮像した画像である必要はなく、たとえば、登録対象を撮像した写真をスキャナ等により読取った画像データや、登録対象をデジタルカメラで撮像した画像データを直接画像処理装置１へ入力して画像情報記憶部１０に登録するように構成してもよい。また、入力画像についても同様に、登録対象を撮像した写真をスキャナ等により読取った画像データや、登録対象をデジタルカメラで撮像した画像データを入力画像として直接画像処理装置１へ入力するように構成してもよい。 In the above embodiment, the image information of the image captured by the image capturing unit 2 is stored and registered in the image information storage unit 10 as a registered image, but the registered image registered in the image information storage unit 10 is The image need not necessarily be an image captured by the imaging unit 2. For example, image data obtained by reading a photograph of the registration target with a scanner or image data obtained by capturing the registration target with a digital camera is directly input to the image processing apparatus 1. Then, the image information storage unit 10 may be registered. Similarly, the input image is configured such that image data obtained by reading a photograph of the registration target with a scanner or image data obtained by capturing the registration target with a digital camera is directly input to the image processing apparatus 1 as an input image. May be.
　また、本実施形態では、本発明に係る画像照合方法の一例として、画像照合装置１による入力画像と登録画像との画像照合方法について説明したが、本発明に係る画像照合方法は、入力画像と登録画像とを照合する画像照合方法に限定するものではなく、第１の画像と第２の画像という２つの任意の画像のうち、少なくともいずれか一方の画像の鮮鋭度を調整して第１の画像と第２の画像との鮮鋭度の差を低減し、該鮮鋭度の差を低減した第１の画像と第２の画像とを照合することに特徴を有するものである。 Further, in the present embodiment, as an example of the image matching method according to the present invention, the image matching method between the input image and the registered image by the image matching device 1 has been described. It is not limited to the image collation method for collating the registered image, and the first image is adjusted by adjusting the sharpness of at least one of the two arbitrary images of the first image and the second image. It is characterized in that the difference in sharpness between the image and the second image is reduced, and the first image and the second image in which the difference in sharpness is reduced are collated.
　かかる本発明の画像照合方法によれば、第１の画像と第２の画像という２つの任意の画像の鮮鋭度の差が画像の照合精度に影響を及ぼす程に異なっていた場合であっても、照合精度の低下度合いを小さくすることができ、第１の画像と第２の画像という任意の２つの画像の撮像対象が同一であるか否かを精度良く判定することができる。 According to the image collating method of the present invention, even if the difference in sharpness between two arbitrary images, ie, the first image and the second image, is different enough to affect the collation accuracy of the image. Thus, the degree of decrease in collation accuracy can be reduced, and it can be accurately determined whether or not the imaging targets of any two images of the first image and the second image are the same.
　ところで、上記した各プログラムについては、必ずしも最初からＲＯＭに記憶させておく必要はなく、例えば、コンピュータに挿入されるフレキシブルディスク（ＦＤ）、ＣＤ－ＲＯＭ、ＭＯディスク、ＤＶＤディスク、光磁気ディスク、ＩＣカードなどの「可搬用の物理媒体」、または、コンピュータの内外に備えられるハードディスクドライブ（ＨＤＤ）などの「固定用の物理媒体」、さらには、公衆回線、インターネット、ＬＡＮ、ＷＡＮなどを介してコンピュータに接続される「他のコンピュータ（またはサーバ）」などに各プログラムを記憶させておき、コンピュータがこれらから各プログラムを読み出して実行するようにしてもよい。 By the way, it is not always necessary to store the above-mentioned programs in the ROM from the beginning. For example, a flexible disk (FD), a CD-ROM, an MO disk, a DVD disk, a magneto-optical disk, an IC inserted into a computer. "Portable physical medium" such as a card, or "fixed physical medium" such as a hard disk drive (HDD) provided inside or outside the computer, and further via a public line, the Internet, LAN, WAN, etc. Each program may be stored in “another computer (or server)” connected to the computer, and the computer may read and execute each program from these.
　また、本実施形態において説明した各処理のうち、自動的に行われるものとして説明した処理の全部または一部を手動的に行うこともでき、あるいは、手動的に行われるものとして説明した処理の全部または一部を公知の方法で自動的におこなうこともできる。この他、上記文書中や図面中で示した処理手順、制御手順、具体的名称、各種の情報やパラメータを含む情報については、特記する場合を除いて任意に変更することができる。 Also, among the processes described in this embodiment, all or part of the processes described as being performed automatically can be performed manually, or the processes described as being performed manually can be performed. All or a part can be automatically performed by a known method. In addition, the processing procedures, control procedures, specific names, information including various types of information and parameters shown in the above-mentioned document and drawings can be arbitrarily changed unless otherwise specified.
　また、図示した各装置の各構成要素は機能概念的なものであり、必ずしも物理的に図示の如く構成されていることを要しない。すなわち、各装置の分散・統合の具体的形態は図示のものに限られず、その全部または一部を、各種の負荷や使用状況などに応じて、任意の単位で機能的または物理的に分散・統合して構成することができる。さらに、各装置にて行なわれる各処理機能は、その全部または任意の一部が、ＣＰＵおよび当該ＣＰＵにて解析実行されるプログラムにて実現され、あるいは、ワイヤードロジックによるハードウェアとして実現され得る。 Also, each component of each illustrated apparatus is functionally conceptual and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured. Further, all or any part of each processing function performed in each device may be realized by a CPU and a program analyzed and executed by the CPU, or may be realized as hardware by wired logic.

Claims (7)

  1. 　画像照合装置であって、
    　登録画像を記憶した画像情報記憶手段と、
    　前記登録画像と照合する入力画像を取得する画像取得手段と、
    　画像の鮮鋭度を検出する鮮鋭度検出手段と、
    　画像の鮮鋭度を調整する鮮鋭度調整手段と、
    　前記鮮鋭度検出手段により前記登録画像及び前記入力画像の鮮鋭度を検出し、前記鮮鋭度調整手段により、少なくとも一方の画像の鮮鋭度を調整して、前記登録画像と前記入力画像との鮮鋭度の差を低減する鮮鋭度差低減手段と、
    　前記鮮鋭度差低減手段により鮮鋭度の差が低減された前記登録画像と前記入力画像とを照合する画像照合手段と、
    　有することを特徴とする画像照合装置。     An image matching device,
    Image information storage means for storing registered images;
    Image acquisition means for acquiring an input image to be collated with the registered image;
    Sharpness detection means for detecting the sharpness of the image;
    Sharpness adjusting means for adjusting the sharpness of the image;
    The sharpness of the registered image and the input image is detected by the sharpness detecting means, the sharpness of at least one of the images is adjusted by the sharpness adjusting means, and the sharpness of the registered image and the input image is adjusted. Sharpness difference reducing means for reducing the difference between,
    An image collating unit that collates the input image with the registered image in which the difference in sharpness is reduced by the sharpness difference reducing unit;
    An image collating apparatus comprising:
  2. 　画像照合装置であって、
    　画像の鮮鋭度を検出する鮮鋭度検出手段と、
    　画像の鮮鋭度を調整する鮮鋭度調整手段と、
    　前記鮮鋭度検出手段により検出される鮮鋭度が所定値に近づくように前記鮮鋭度調整手段によって調整された画像を登録画像として記憶した画像情報記憶手段と、
    　前記登録画像と照合する入力画像を取得する画像取得手段と、
    　前記画像取得手段により取得された入力画像の鮮鋭度を前記所定値に近づけるように調整して、該入力画像と前記登録画像との鮮鋭度の差を低減する鮮鋭度差低減手段と、
    　前記鮮鋭度差低減手段により鮮鋭度の差が低減された前記入力画像と前記登録画像とを照合する画像照合手段と、
    　を有することを特徴とする画像照合装置。     An image matching device,
    Sharpness detection means for detecting the sharpness of the image;
    Sharpness adjusting means for adjusting the sharpness of the image;
    Image information storage means for storing an image adjusted by the sharpness adjusting means so that the sharpness detected by the sharpness detecting means approaches a predetermined value, as a registered image;
    Image acquisition means for acquiring an input image to be collated with the registered image;
    A sharpness difference reducing means for adjusting the sharpness of the input image acquired by the image acquisition means so as to approach the predetermined value, and reducing a difference in the sharpness between the input image and the registered image;
    Image collating means for collating the input image and the registered image whose sharpness difference is reduced by the sharpness difference reducing means;
    An image collating apparatus comprising:
  3. 　画像照合装置であって、
    　画像の鮮鋭度を検出する鮮鋭度検出手段と、
    　画像の鮮鋭度を調整する鮮鋭度調整手段と、
    　前記鮮鋭度検出手段により検出される鮮鋭度が所定値に近づくように前記鮮鋭度調整手段によって調整された登録画像より特徴を抽出した登録特徴情報を記憶した画像情報記憶手段と、
    　前記登録画像と照合する入力画像を取得する画像取得手段と、
    　前記画像取得手段により取得された前記入力画像の鮮鋭度を前記所定値に近づけるように調整して、前記入力画像と前記登録画像との鮮鋭度の差を低減する鮮鋭度差低減手段と、
    　前記鮮鋭度差低減手段により鮮鋭度が調整された前記入力画像の特徴を抽出した入力特徴情報と前記登録特徴情報とを照合することにより前記入力画像と前記登録画像とを照合する画像照合手段と、
    　を有することを特徴とする画像照合装置。     An image matching device,
    Sharpness detection means for detecting the sharpness of the image;
    Sharpness adjusting means for adjusting the sharpness of the image;
    Image information storage means for storing registered feature information obtained by extracting features from the registered image adjusted by the sharpness adjusting means so that the sharpness detected by the sharpness detecting means approaches a predetermined value;
    Image acquisition means for acquiring an input image to be collated with the registered image;
    A sharpness difference reducing means for adjusting the sharpness of the input image acquired by the image acquisition means so as to approach the predetermined value, and reducing a difference in sharpness between the input image and the registered image;
    Image collating means for collating the input image with the registered image by collating input feature information obtained by extracting features of the input image whose sharpness has been adjusted by the sharpness difference reducing unit and the registered feature information; ,
    An image collating apparatus comprising:
  4. 　前記画像照合手段は、前記登録画像と前記入力画像とを照合して、前記登録画像と前記入力画像との類似度を算出することを特徴とする請求項１～３のいずれか１項に記載の画像照合装置。 The image collating unit collates the registered image with the input image, and calculates a similarity between the registered image and the input image. Image matching device.
  5. 　前記画像情報記憶手段から、前記入力画像と照合する登録画像に対応した情報を選択する照合対象選択手段をさらに備え、
    　前記画像照合手段は、前記照合対象選択手段により選択された情報を用いて、前記入力画像の撮像対象と前記登録画像の撮像対象とが同一であるか否かを判定することを特徴とする請求項１～３のいずれか１項に記載の画像照合装置。     The image information storage means further comprises a collation target selection means for selecting information corresponding to a registered image to be collated with the input image,
    The image collating unit determines whether the imaging target of the input image and the imaging target of the registered image are the same using the information selected by the collation target selecting unit. Item 4. The image collating device according to any one of Items 1 to 3.
  6. 　前記鮮鋭度差低減手段は、前記入力画像と前記登録画像との鮮鋭度の差が所定の差よりも大きいときに、前記入力画像と前記登録画像との鮮鋭度の差を低減することを特徴とする請求項１～３のいずれか１項に記載の画像照合装置。 The sharpness difference reducing means reduces the difference in sharpness between the input image and the registered image when the difference in sharpness between the input image and the registered image is larger than a predetermined difference. The image collating apparatus according to any one of claims 1 to 3.
  7. 　画像照合方法であって、
    　第１の画像と第２の画像とのうち少なくとも一方の画像の鮮鋭度を調整して、前記第１の画像と第２の画像との鮮鋭度の差を低減するステップと、
    　鮮鋭度の差を低減した前記第１の画像と第２の画像とを照合するステップと、
    　を有することを特徴とする画像照合方法。     An image matching method,
    Adjusting the sharpness of at least one of the first image and the second image to reduce the difference in sharpness between the first image and the second image;
    Collating the first and second images with reduced sharpness differences;
    An image matching method characterized by comprising:
PCT/JP2008/063976 2008-08-04 2008-08-04 Image collating unit and image collating method WO2010016103A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP2010523662A JP5395795B2 (en) 2008-08-04 2008-08-04 Image collation apparatus and image collation method
PCT/JP2008/063976 WO2010016103A1 (en) 2008-08-04 2008-08-04 Image collating unit and image collating method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2008/063976 WO2010016103A1 (en) 2008-08-04 2008-08-04 Image collating unit and image collating method

Publications (1)

Publication Number Publication Date
WO2010016103A1 true WO2010016103A1 (en) 2010-02-11

Family

ID=41663333

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2008/063976 WO2010016103A1 (en) 2008-08-04 2008-08-04 Image collating unit and image collating method

Country Status (2)

Country Link
JP (1) JP5395795B2 (en)
WO (1) WO2010016103A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2502817A (en) * 2012-06-08 2013-12-11 Siemens Medical Solutions Improving comparison between scan and archive images
JP2016530593A (en) * 2013-06-04 2016-09-29 ペイパル インコーポレイテッド Assessing the clarity of an image

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH04123276A (en) * 1990-09-14 1992-04-23 Fujitsu Ltd Device for collating finger print
JP2003324610A (en) * 2002-02-27 2003-11-14 Konica Minolta Holdings Inc Image processing method and image converting equipment
JP2006072555A (en) * 2004-08-31 2006-03-16 Secom Co Ltd Biological information image correcting device

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH04123276A (en) * 1990-09-14 1992-04-23 Fujitsu Ltd Device for collating finger print
JP2003324610A (en) * 2002-02-27 2003-11-14 Konica Minolta Holdings Inc Image processing method and image converting equipment
JP2006072555A (en) * 2004-08-31 2006-03-16 Secom Co Ltd Biological information image correcting device

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2502817A (en) * 2012-06-08 2013-12-11 Siemens Medical Solutions Improving comparison between scan and archive images
US9299144B2 (en) 2012-06-08 2016-03-29 Siemens Medical Solutions Usa, Inc. Equalizing smoothing for a scan comparison to database
GB2502817B (en) * 2012-06-08 2019-11-06 Siemens Medical Solutions Usa Inc Equalising smoothing for scan comparison to database
JP2016530593A (en) * 2013-06-04 2016-09-29 ペイパル インコーポレイテッド Assessing the clarity of an image

Also Published As

Publication number Publication date
JPWO2010016103A1 (en) 2012-01-12
JP5395795B2 (en) 2014-01-22

Similar Documents

Publication Publication Date Title
US6535636B1 (en) Method for automatically detecting digital images that are undesirable for placing in albums
US8306288B2 (en) Automatic identification of fingerprint inpainting target areas
US7729536B2 (en) Boundary extracting method, program, and device using the same
US10679358B2 (en) Learning image automatic sorting device, learning image automatic sorting method, and learning image automatic sorting program
US20110038545A1 (en) Scale robust feature-based identifiers for image identification
JP4518139B2 (en) Image processing device
WO2008140539A1 (en) Methods for gray-level ridge feature extraction and associated print matching
RU2745098C2 (en) Method of determinig authorship of a painting
CN110728201B (en) Image processing method and device for fingerprint identification
JP5395795B2 (en) Image collation apparatus and image collation method
JP2003337949A (en) Fingerprint matching device
JP5179329B2 (en) Face recognition system
CN112101296A (en) Face registration method, face verification method, device and system
KR100888674B1 (en) Method for measuring similarity using frenquency range
EP1638038A1 (en) Method of determining an image descriptor, image comparison method, image search method and associated devices
CN112183454A (en) Image detection method and device, storage medium and terminal
JP4812743B2 (en) Face recognition device, face recognition method, face recognition program, and recording medium recording the program
CN117474916B (en) Image detection method, electronic equipment and storage medium
EP4202857A1 (en) Texture-based authentication of digital identity documents
Neubert et al. Image pre-processing detection: Evaluation of Benford's law, spatial and frequency domain feature performance
CN114463792B (en) Multispectral identification method, multispectral identification device, multispectral identification equipment and readable storage medium
CN108780493B (en) Method and device for detecting wet hand area in fingerprint image
CN108288262B (en) Power coil wire looseness detection method and device, computer readable storage medium and terminal equipment
Fan et al. Feature based automatic stitching of microscopic images
WO2007108056A1 (en) Video image recognition system and video image recognition program

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: 08792173

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2010523662

Country of ref document: JP

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 08792173

Country of ref document: EP

Kind code of ref document: A1