JP5773737B2 - Verification device - Google Patents

Description

  The present invention relates to a collation device that collates a reference image and a registered image.

  In recent years, in the fields that require personal authentication, such as access control to computer rooms and important machine rooms, and access control to computer terminals and bank financial terminals, biometric information has been used instead of conventional PINs and ID cards. A system for personal authentication based on the above has been developed.

  Personal authentication systems using biometric information are attracting attention because they do not require memory or possession and have a low risk of theft. In personal authentication using biometric information, fingerprints, irises, faces, palm prints, handwriting, and the like are used. Among them, personal authentication using a face is attracting attention because of its high convenience and acceptability.

  For example, in a personal authentication system used for entrance / exit management of a building or the like, the face of a person who wants to enter a room is captured as a reference image, and this reference image and a pre-registered person's face image (registered image) If the reference image matches the registered image, the person is allowed to enter the room.

  In the collation process between the reference image and the registered image, the applicant assigns, for each of the reference image and the registered image, the original image as a high-resolution image with the highest resolution, and the resolution with the high-resolution image as the uppermost layer. A plurality of different hierarchical images are created, and corresponding points are searched from the low-resolution image to the high-resolution image of the hierarchical image using a phase-only correlation (POC) method (for example, Patent Document 1). 2), the reference image is registered with the registered image based on the phase-only correlation value of the corresponding region after the search. Hereinafter, this collation method using the corresponding point search using the hierarchical image is referred to as a hierarchical POC.

  In this hierarchical POC, for example, for the reference image, the original image (128 × 128 pixel image) is set as a high resolution image, and an image of 64 × 64 pixels (first low resolution image) is formed below this high resolution image. ) And a 32 × 32 pixel image (second low-resolution image) is created below the first low-resolution image. As for the registered image, as in the reference image, the original image (128 × 128 pixel image) is used as a high resolution image, and an image of 64 × 64 pixels (first low resolution) is formed below this high resolution image. Image), and an image of 32 × 32 pixels (second low-resolution image) is created below the first low-resolution image. In the hierarchical image including the original image, the layer of the high resolution image is the 0th layer, the layer of the first low resolution image is the first layer, and the layer of the second low resolution image is the second layer.

  Then, m reference points in the x-axis direction and n reference points in the y-axis direction are determined with respect to the original image of the registered image, and the lowest layer (second layer) is set for each reference point determined in the original image of the registered image. A search window (local area) is determined from the hierarchical image of and the search for the corresponding area in the reference image corresponding to the local area is started. The search for the corresponding area is performed by POC collation, and when the search for the corresponding area is completed for the hierarchical image, the search result for the hierarchical image is reflected in the search for the next hierarchy (first layer). Then, the corresponding area is sequentially searched for the hierarchical image having a high resolution, and the search for the corresponding area is repeated until the hierarchical image of the uppermost layer (0th layer) is reached.

  If a corresponding region is found in the uppermost layer (0th layer) hierarchical image, the phase-only correlation value of the corresponding region is obtained. Hereinafter, similarly, for each local region of the registered image, the corresponding region of the reference image is searched from the low-resolution image to the high-resolution image, and the phase-only correlation value of the corresponding region after the search is obtained. The reference image and the registered image are collated based on the limited correlation value.

  Note that POC mathematically processes the original image data and the image data to be collated with Fourier transform, and decomposes the data into amplitude (shading data) and phase (image contour data). Of these, only the phase information is used to obtain the correlation between the two images, which is characterized by being resistant to disturbances and capable of obtaining a highly accurate calculation result.

  The details of the POC are described in Patent Document 3 as well as Patent Documents 1 and 2 described above. In POC, one image is subjected to two-dimensional discrete Fourier transform to create a Fourier image. Then, a Fourier image of the other image created by performing the same processing as this Fourier image is synthesized, and a processing for setting the amplitude to 1 is performed on this synthesized Fourier image, and then a two-dimensional discrete Fourier Converted into correlation data. The correlation data (POC function) has an amplitude of 1 in the frequency space and is only in phase, but can be considered as data obtained by convolving one image with the other image. This represents the correlation between the first image and the other image.

JP 2008-209275 A JP 2008-123141 A Japanese Patent Laid-Open No. 10-63847

  However, in the above-described collation using the hierarchical POC, the reference image and the registered image can be collated with high accuracy. However, when applied to 1: N collation in which the reference image is collated with a plurality of registered images, processing is performed. There is a problem that the amount increases and the collation speed decreases.

  The present invention has been made to solve such problems, and an object of the present invention is to provide a collation apparatus capable of achieving both high accuracy and high speed processing.

In order to achieve such an object, the present invention provides a collation apparatus that collates a reference image with a registered image. For a reference image, the original image is a high-resolution image with the highest resolution, and the high-resolution image is the top layer. and to a hierarchical image creation means for the high-resolution image resolution lowered only resolution in the same image area and to create a hierarchy images of different low-resolution, storing a plurality of registered images of the reference image and the same resolution Registered image storage means, a registered image reading means for reading out the registered images stored in the registered image storage means one by one, a low-resolution hierarchical image of the reference image created by the hierarchical image creating means, and a registered image reading means any low-resolution hierarchy image calculation of similarity between the same low-resolution hierarchy image among the layer image corresponding to the hierarchical image of the reference image of the read registered images by Done from, based on the calculated similarity, characterized in that it comprises a first registration image candidate selecting means for selecting as the registration image candidates to be matched with the reference image registration image read by the registered image reading unit And

In the present invention, when the reference image and the registered image are collated, the reference image is defined as a high-resolution image having the highest resolution, and the plurality of low-resolution layers having different resolutions with the high-resolution image as the uppermost layer. Create an image. Then, the registered images are read one by one from the registered image storage unit, and it is determined whether or not the read registered images are to be registered image candidates to be compared with the reference image. That is, before the reference image and the registered image are collated, the registered images to be actually collated are narrowed down. As a result, the number of collations in 1: N collation is reduced, and both high accuracy and high-speed processing are achieved.

  In the present invention, the registered image storage means stores a plurality of registered images having the same resolution as the reference image. In this case, as the registered image, only the original image of the registered image may be stored, or a hierarchical image including the original image of the registered image may be stored. When storing a hierarchical image of a registered image, a plurality of hierarchical images having different resolutions, the original image being the highest resolution high resolution image and the highest resolution image corresponding to the reference image hierarchical image. Try to remember.

  In the present invention, when a hierarchical image of a registered image is stored, a hierarchical image including the original image is read as the registered image. Then, based on the similarity between the created low-resolution hierarchical image of the reference image and the same low-resolution hierarchical image among the hierarchical images corresponding to the read-out hierarchical image of the registered image, the registered image A candidate is selected.

  In the present invention, when only the original image of the registered image is stored, the original image is read as the registered image, this original image is set as the highest resolution high resolution image, and this high resolution image is set as the uppermost layer. A plurality of hierarchical images having different resolutions are created. The hierarchical image of the registered image is created corresponding to the hierarchical image of the reference image. Then, based on the similarity between the low-resolution hierarchical image of the created reference image and the same low-resolution hierarchical image among the hierarchical images corresponding to the hierarchical image of the created reference image of the registered image, registered image candidates Is selected.

In the present invention, the calculation of the similarity of the same low-resolution hierarchy image, to perform a low-resolution hierarchy image of arbitrary. In this case, for example, when the calculated similarity satisfies a predetermined reference value, it is selected as a registered image candidate of the read registered image. In this case, for example, a reference value determined according to the resolution of the hierarchy is used as the reference value for each hierarchy of the hierarchy image.

  In the present invention, measurement information acquisition means for acquiring predetermined measurement information from a reference image, and reference from a plurality of registered images stored in the registered image storage means based on the measurement information acquired by the measurement information acquisition means A second registered image candidate selecting unit that selects a registered image candidate to be collated with the image, and the first registered image candidate selecting unit combines the registered image candidates selected by the second registered image candidate selecting unit. A configuration may be adopted in which registered image candidates are selected.

  For example, the pupil distance, face size, pupil size, face curvature, etc. are acquired as predetermined measurement information from the reference image, and the registered images are narrowed down based on the acquired measurement information. A final registered image candidate is obtained by combining the registered image narrowed down based on the measurement information and the registered image narrowed down based on the similarity of the same low-resolution hierarchical image.

  In this case, the registered image narrowed down based on the measurement information is further narrowed down based on the similarity of the same low-resolution hierarchical image to obtain a final registered image candidate, or the registered image narrowed down based on the measurement information AND with the registered image narrowed down based on the similarity of the same low-resolution hierarchical image to obtain the final registered image candidate, or the same low resolution as the registered image narrowed down based on the measurement information Various schemes are conceivable, such as obtaining a final registered image candidate by ORing with the registered images narrowed down based on the similarity of the layer images.

  In the present invention, either one of the hierarchical image of the reference image created by the hierarchical image creating unit and the hierarchical image corresponding to the hierarchical image of the reference image of the registered image read by the registered image reading unit is previously stored. Feature area setting means for setting a predetermined feature area, and corresponding area search means for searching a corresponding area corresponding to the feature area of one hierarchical image for which the characteristic area is set by the characteristic area setting means from the other hierarchical image Based on the similarity between the feature region set in one hierarchical image by the feature region setting unit and the corresponding region searched from the other hierarchical image by the corresponding region searching unit. Second registered image candidate selecting means for selecting a registered image as a registered image candidate to be compared with a reference image, and the first registered image candidate selecting means It may be configured to select the registration image candidates by combining the registration image candidate selected by the second registration image candidate selection unit.

  In this case, a feature region is set for the hierarchical image of the reference image or the hierarchical image of the registered image. For example, a zone including characteristic portions such as eyes and noses is set as the characteristic region. Then, the corresponding area corresponding to the feature area of the one hierarchical image in which the feature area is set is searched from the other hierarchical image, and the correspondence area searched from the characteristic area set in the one hierarchical image and the other hierarchical image. A registered image candidate is selected based on the similarity with the region.

  In the present invention, a second registered image candidate selecting unit that selects a registered image candidate to be compared with a reference image from a plurality of registered images stored in the registered image storage unit based on the acquired measurement information. A feature region predetermined for any one of the hierarchical image of the reference image created by the hierarchical image creating means and the hierarchical image corresponding to the hierarchical image of the reference image of the registered image read by the registered image reading means A feature area setting means for setting the feature area, a corresponding area search means for searching for a corresponding area corresponding to the feature area of the one hierarchical image in which the feature area is set by the feature area setting means, and a feature area setting Based on the similarity between the feature region set in one hierarchical image by the means and the corresponding region searched from the other hierarchical image by the corresponding region search means, A third registered image candidate selecting unit that selects the registered image read by the recorded image reading unit as a registered image candidate to be collated with the reference image, and the first registered image candidate selecting unit includes the second registered image; The registered image candidate may be selected by combining the registered image candidate selected by the candidate selecting unit and the registered image candidate selected by the third registered image candidate selecting unit.

  In the present invention, when the number of selected registered image candidates is 0, the result of matching between the reference image and the registered image may be determined to be inconsistent. For example, if the number of selected registered image candidates is 0 even though all stored registered images have been read, the collation result between the reference image and the registered image is made inconsistent. In this way, the collation result can be obtained before collating the reference image with the registered image, that is, at the stage of the classification process for narrowing down the registered image to be collated with the reference image.

  According to the present invention, the registered images are read one by one, and the similarity between the low-resolution hierarchical image of the reference image and the same low-resolution hierarchical image among the hierarchical images corresponding to the hierarchical image of the reference image of the registered image is determined. Based on this, since the read registered image is selected as a registered image candidate to be collated with the reference image, the registered image to be collated with the reference image is narrowed down before collating the reference image with the registered image. Thus, the number of collations in 1: N collation can be reduced, and both high accuracy and high speed processing can be achieved.

It is a block block diagram of the collation apparatus which shows one embodiment of this invention. It is a flowchart which shows the registration process of the target object in this collation apparatus. It is a flowchart which shows the collation process of the target object in this collation apparatus. It is a flowchart which shows the 1st example (selection process 1 of a registration image candidate) of the selection process of the registration image candidate in the collation process of the target object in this collation apparatus. It is a figure which illustrates the hierarchical image of the reference image produced by the selection process of the registration image candidate, and the hierarchical image of the read registered image. It is a flowchart which shows the 2nd example (selection process 2 of a registration image candidate) of the selection process of a registration image candidate. 12 is a flowchart of registered image processing 3 used in combination with registered image candidate selection processing 1 and 2. 10 is a flowchart of registered image processing 4 used in combination with registered image candidate selection processing 1 and 2. It is a figure which illustrates the hierarchical image of the reference image created by the selection process 4 of the registered image candidate, and the hierarchical image of the read registered image together with the feature region and the corresponding region. It is a figure which shows the characteristic area set with respect to the reference | standard point previously defined as a point which prescribes | regulates a characteristic area, and the uppermost hierarchy image (high resolution image (original image)) of a registration image centering on this reference point . 12 is a flowchart illustrating a modification of registered image candidate selection processing 1; 12 is a flowchart illustrating a modification of a registered image candidate selection process 2; 15 is a flowchart illustrating a modification of registered image candidate selection processing 4;

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a block diagram of a collation apparatus showing an embodiment of the present invention. In the figure, 10 is a first CCD camera, 11 is a second CCD camera, 12 is a liquid crystal display (LCD), 20 is a processing unit, and the processing unit 20 includes a control unit 20-1 having a CPU. ROM 20-2, RAM 20-3, hard disk (HD) 20-4, frame memory (FM) 20-5, external connection unit (I / F) 20-6, and Fourier transform unit (FFT) 20 -7, and a registration program and a verification program are stored in the ROM 20-2. The CCD cameras 10 and 11 are installed with a predetermined distance L apart. That is, the CCD cameras 10 and 11 are arranged side by side in the horizontal direction with the distance between the lenses 10-1 and 11-1 being L. For easy understanding in the figure, the collation device is a diagram seen from above, and the object (human face) is a diagram seen from the side.

[Registration of objects]
In this collation device, when the object is a human face M1, registration of the object (registered object) M1 is performed as follows. Before operation, the user turns the face M1 toward the CCD cameras 10 and 11 and presses a registration start switch (not shown).

[Image capture]
When the registration start switch is pressed (FIG. 2: YES in step S101), the control unit 20-1 displays an image capturing the registration object M1 from the CCD camera 10 via the frame memory 20-5. An image captured as an image and capturing the registered object M1 from the CCD camera 11 is captured as an image of the right camera (step S102).

[Detection of face area]
Then, the control unit 20-1 detects a face area by image processing from the captured left camera image and right camera image (step S103). At this time, if the brightness of the face area is not within a certain range, gain control is performed, and the left camera image and the right camera image are captured again. In this example, the left and right face areas are detected, but either the left or right face area may be detected.

[Detection of pupil region and normalization of face image]
Then, the control unit 20-1 detects the pupil region by image processing from the detected left and right face region images (left and right face images) (step S104), and the distance R from the camera to the registration target M1 by 3D measurement. Is calculated (step S105), and the size of the right face image is normalized based on the calculated distance R (step S106).

(Create hierarchical image)
Then, the control unit 20-1 uses the face image normalized in step S106 as the original image of the registered image, and sets the original image (128 × 128 pixel image) as the high-resolution image of the uppermost layer (0th layer). An image of 64 × 64 pixels (first low resolution image) is created in the lower layer (first layer) of the high resolution image, and 32 × 32 is formed in the lower layer (second layer) of the first low resolution image. An image of the pixel (second low-resolution image) is created (step S107).

  Then, the control unit 20-1 stores the hierarchical image (the high resolution image, the first low resolution image, and the second low resolution image) including the original image in the HD 20-4 as a registered image of the user (Step S1). S108). Similarly, for all users, the registered images (hierarchical images including the original image) of the users are stored in the HD 20-4. In this embodiment, the number of users is set to several thousands, and registered images of these users are stored in the HD 20-4.

  In this example, the right face image is registered, but the left face image may be registered, or the left and right face images may be registered. In this embodiment, for the sake of explanation, the right face image is registered.

[Verification of objects]
In this collation device, when the collation target is the human face M1, the collation of the face M1 is performed as follows. FIG. 3 shows a flowchart of collation processing executed by the control unit 20-1 according to the collation program stored in the ROM 20-2.

[Image capture]
The user stands in front of the CCD cameras 10 and 11. Then, the control unit 20-1 recognizes that the face M1 has entered the imaging area of the CCD cameras 10 and 11, and displays an image capturing the face M1 from the CCD camera 10 via the frame memory 20-5. A camera image is captured, and an image capturing the face M1 from the CCD camera 11 is captured as a right camera image (step S201).

[Detection of face area]
Then, the control unit 20-1 detects a face area by image processing from the captured left camera image and right camera image (step S202). At this time, if the brightness of the face area is not within a certain range, gain control is performed, and the left camera image and the right camera image are captured again. In this example, the left and right face areas are detected, but either the left or right face area may be detected.

[Detection of pupil region and normalization of face image]
Then, the control unit 20-1 detects the pupil region by image processing from the detected left and right face region images (left and right face images) (step S203), and the distance R from the camera to the verification target M1 by 3D measurement. Is calculated (step S204), and the size of the right face image is normalized based on the calculated distance R (step S205).

[Registration Image Candidate Selection: Registration Image Candidate Selection Processing 1 (Embodiment 1)]
Then, the control unit 20-1 uses the face image normalized in step S205 as the original image of the reference image, and starts registration image candidate selection processing to be compared with the reference image (step S206). FIG. 4 shows a flowchart of the first example of the registered image candidate selection process (registered image candidate selection process 1) as the first embodiment.

  In this registered image candidate selection process 1, the control unit 20-1 sets the original image (128 × 128 pixel image) of the reference image obtained in step S205 as a high-resolution image of the uppermost layer (0th layer). An image of 64 × 64 pixels (first low resolution image) is created in the lower layer (first layer) of the high resolution image, and 32 × 32 is formed in the lower layer (second layer) of the first low resolution image. A pixel image (second low-resolution image) is created (step S301).

  Then, the control unit 20-1 reads the first registered image (hierarchical image including the original image) stored in the HD 20-4 (step S302). That is, for the first registered image, a high-resolution image (128 × 128 pixel image) located in the uppermost layer (0th layer) and a first layer located in the lower layer (first layer) of this high-resolution image. Low resolution image (64 × 64 pixel image) and a second low resolution image (32 × 32 pixel image) located in the lower layer (second layer) of the first low resolution image. .

  FIGS. 5A and 5B illustrate the hierarchical image of the reference image created in step S301 and the hierarchical image of the registered image read out in step S302. In FIG. 5A, J0 is a hierarchical image (high-resolution image (128 × 128 pixel image)) located in the 0th layer of the registered image, and J1 is a hierarchical image (first image) located in the 1st layer of the registered image. 1 is a low-resolution image (64 × 64 pixel image)), J2 is a hierarchical image (second low-resolution image (32 × 32 pixel image)) located in the second layer of the registered image, In FIG. 5B, I0 is a hierarchical image located in the 0th layer of the reference image (high resolution image (128 × 128 pixel image)), and I1 is a hierarchical image located in the 1st layer of the reference image (first image). I2 is a hierarchical image (second low-resolution image (32 × 32 pixel image)) located in the second layer of the reference image.

  The control unit 20-1 obtains the hierarchical images I0, I1, and I2 of the reference image and the hierarchical images J0, J1, and J2 of the registered image, and then sets N = 2 (step S303). The similarity between the hierarchical image of the layer and the hierarchical image of the Nth layer of the registered image is calculated (step S304). In this case, since N = 2, the similarity between the second layer hierarchical image I2 of the reference image and the second layer hierarchical image J2 of the registered image is calculated.

  Then, a predetermined reference value TH of the second layer is read (step S305), and the calculated similarity is compared with the read reference value TH of the second layer (step S306). Here, the reference value TH of the second layer is determined according to the resolution of the layer.

  If the calculated similarity is greater than or equal to the reference value TH of the second layer (YES in step S306), the registered image read in step S302 is selected as a registered image candidate (step S307), and the number of selected registered image candidates M (initial value 0) is set to M = M + 1 (step S308).

  Then, after confirming that the number M of the selected registered image candidates has not reached 100 (NO in step S309), confirm that other registered images are stored in the HD 20-4 (step S309). (YES in S312), the next registered image is read out from the HD 20-4 (step S302), and the processes in and after step S303 are repeated.

  If the calculated similarity is not equal to or higher than the reference value TH of the second layer (NO in step S306), it is confirmed that there is a hierarchical image with a higher resolution (YES in step S310), and N = N−1 ( Step S311) and return to Step S304. In this case, since N = 1, the control unit 20-1 calculates the similarity between the first layer hierarchical image I1 of the reference image and the first layer hierarchical image J1 of the registered image.

  Then, a predetermined reference value TH of the first layer is read (step S305), and the calculated similarity is compared with the read reference value TH of the first layer (step S306). Here, the reference value TH of the first layer is determined according to the resolution of the layer.

  If the calculated similarity is greater than or equal to the first layer reference value TH (YES in step S306), the registered image read in step S302 is selected as a registered image candidate (step S307), and the number of selected registered image candidates M is set to M = M + 1 (step S308).

  Then, after confirming that the number M of the selected registered image candidates has not reached 100 (NO in step S309), confirm that other registered images are stored in the HD 20-4 (step S309). (YES in S312), the next registered image is read out from the HD 20-4 (step S302), and the processes in and after step S303 are repeated.

  If the calculated similarity is not equal to or higher than the reference value TH of the first layer (NO in step S306), it is confirmed that there is a hierarchical image with a higher resolution (YES in step S310), and N = N−1 ( Step S311) and return to Step S304. In this case, since N = 0, the control unit 20-1 calculates the similarity between the 0th layer image I0 of the reference image and the 0th layer image J0 of the registered image.

  Then, a predetermined reference value TH for the 0th layer is read (step S305), and the calculated similarity is compared with the read reference value TH for the 0th layer (step S306). Here, the reference value TH of the 0th layer is determined according to the resolution of the layer.

  If the calculated similarity is greater than or equal to the 0th layer reference value TH (YES in step S306), the registered image read in step S302 is selected as a registered image candidate (step S307), and the number of selected registered image candidates is selected. M is set to M = M + 1 (step S308).

  Then, after confirming that the number M of the selected registered image candidates has not reached 100 (NO in step S309), confirm that other registered images are stored in the HD 20-4 (step S309). (YES in S312), the next registered image is read out from the HD 20-4 (step S302), and the processes in and after step S303 are repeated.

  If the calculated similarity is not equal to or higher than the reference value TH of the 0th layer (NO in step S306), the control unit 20-1 checks whether there is a hierarchical image with a higher resolution (YES in step S310). . In this case, since there is no hierarchical image with a resolution higher than that of the 0th layer (NO in step S310), after confirming that other registered images are stored in the HD 20-4 (YES in step S312), the HD 20- The next registered image is read from 4 (step S302), and the processing from step S303 is repeated.

  As a result, the hierarchical images of the registered images are read from the HD 20-4 one after another, and the similarity between the hierarchical image of the registered image and the hierarchical image of the reference image is calculated in ascending order of resolution, and the calculated similarity and its hierarchy are calculated. Are compared with the reference value TH, and a registered image whose similarity is confirmed to be equal to or higher than the reference value TH in any hierarchy is selected as a registered image candidate.

  When the number M of selected registered image candidates becomes 100 during the repetition of this process (YES in step S309), the control unit 20-1 ends the registered image candidate selection process, and is shown in FIG. The process proceeds to step S207.

  Note that during the repetition of this process, if there is no other registered image before the number M of selected registered image candidates reaches 100 (NO in step S313), the control unit 20-1 selects the selected registered image. It is checked whether there is a candidate (step S313). If there is a selected registered image candidate (YES in step S313), the registered image candidate selection process is terminated, and the process proceeds to step S207 shown in FIG.

  On the other hand, if there is no selected registered image candidate (NO in step S313), that is, the number of selected registered image candidates is 0 even though all the registered images are read from the HD 20-4. If there is, the collation result between the reference image and the registered image is set to be inconsistent (step S314), and the process is terminated. The collation result in this case is displayed on the liquid crystal display device 12.

[Verification of reference image and registered image candidate]
After the registration image candidate selection process described above (step S206), the control unit 20-1 reads the first registered image candidate selected by the selection process from the HD 20-4 (step S207). Then, the reference image obtained in step S205 and the registered image candidate read in step S207 are collated with the hierarchical POC (step S208). Since the hierarchical POC has been described above, a detailed description thereof is omitted here.

  Then, the correlation value obtained in this hierarchy POC is compared with a threshold value (step S209), and if the correlation value is equal to or greater than the threshold value (YES in step S209), the collation result between the reference image and the registered image is made coincident (step S210). ), The process is terminated. The collation result in this case is displayed on the liquid crystal display device 12. Moreover, it outputs as an unlocking command etc. via the external connection part 20-6.

  If the correlation value is not greater than or equal to the threshold value (NO in step S209), the control unit 20-1 confirms that other registered image candidates are stored in the HD 20-4 (YES in step S211), and then HD20- The next registered image candidate is read from 4 (step S207), and the processing from step S208 is repeated.

  As a result, the registered image candidates are successively read from the HD 20-4 until a matching result that matches is obtained, and matching in the hierarchical POC between the read registered image candidates and the reference image is repeated.

  If there is no other registered image candidate during the repetition of the collation at this hierarchical POC (NO in step S211), the control unit 20-1 determines that the collation result between the reference image and the registered image does not match (step S212). Exit. The collation result in this case is displayed on the liquid crystal display device 12.

  In the first embodiment, N = 2 is set in step S303, and the similar calculation of the registered image and the reference image is started from the second layer (lowermost layer) hierarchical image. However, in step S303, N = 1 and calculation of similarities between the registered image and the reference image may be started from the first layer image.

  In this way, in the present embodiment, the registered image to be compared with the reference image is narrowed down before the reference image and the registered image are collated, and the number of collations in 1: N collation is reduced and high accuracy is achieved. High-speed processing is compatible.

  In this example, the registered image candidate and the reference image are collated with the hierarchical POC. However, it is not necessarily performed with the hierarchical POC, and a general collation method using only the original image is adopted. May be.

  In this example, a two-dimensional image is collated, but a three-dimensional image may be collated. When collating three-dimensional images, it is conceivable to apply a collation method based on an ICP (Iterative Closest Point) algorithm.

[Embodiment 2: Registration Image Candidate Selection Process 2]
FIG. 6 shows another example of registered image candidate selection processing (registered image candidate selection processing 2) in step 206 in the flowchart shown in FIG.

  In the registration image candidate selection process 2, the control unit 20-1 sets the original image (128 × 128 pixel image) of the reference image obtained in step S205 as a high-resolution image of the uppermost layer (0th layer). An image of 64 × 64 pixels (first low resolution image) is created in the lower layer (first layer) of the high resolution image, and 32 × 32 is formed in the lower layer (second layer) of the first low resolution image. A pixel image (second low-resolution image) is created (see step S401, FIG. 5B).

  Then, the control unit 20-1 sets N = 2 (step S402), and reads the first registered image (hierarchical image including the original image) stored in the HD 20-4 (step S403). That is, for the first registered image, a high-resolution image (128 × 128 pixel image) located in the uppermost layer (0th layer) and a first layer located in the lower layer (first layer) of this high-resolution image. Low resolution image (64 × 64 pixel image) and a second low resolution image (32 × 32 pixel image) located in the lower layer (second layer) of the first low resolution image. (See FIG. 5 (a)).

  Next, the control unit 20-1 calculates the similarity between the N-th layer image of the reference image and the N-th layer image of the registered image (step S404). In this case, since N = 2, the similarity between the second layer hierarchical image I2 of the reference image and the second layer hierarchical image J2 of the registered image is calculated.

  Then, a predetermined second layer reference value TH is read (step S405), and the calculated similarity is compared with the read second layer reference value TH (step S406). Here, the reference value TH of the second layer is determined according to the resolution of the layer.

  If the calculated similarity is equal to or higher than the second layer reference value TH (YES in step S406), the registered image read in step S302 is selected as a registered image candidate (step S407), and the number of selected registered image candidates is selected. M (initial value 0) is set to M = M + 1 (step S408).

  Then, after confirming that the number M of selected registered image candidates has not reached 100 (NO in step S409), confirm that other registered images are stored in the HD 20-4 (step S409). (YES in S410), the next registered image is read from the HD 20-4 (step S403), and the processes in and after step S404 are repeated.

  If the calculated similarity is not equal to or higher than the second layer reference value TH (NO in step S406), it is confirmed that another registered image is stored in the HD 20-4 (YES in step S410), and then the HD 20- 4 reads the next registered image (step S403), and repeats the processing from step S404.

  Thereby, the hierarchical images of the registered images are read from the HD 20-4 one after another, and the similarity between the second hierarchical image of the registered image and the second hierarchical image of the reference image is calculated and calculated. The similarity is compared with the second layer reference value TH, and a registered image whose calculated similarity is confirmed to be equal to or higher than the reference value TH is selected as a registered image candidate.

  During this process, when all registered images are read from the HD 20-4 and there are no other registered images (NO in step S410), the control unit 20-1 has a hierarchical image with higher resolution. (YES in step S411), N = N-1 is set (step S412), and the process returns to step S403.

  As a result, the control unit 20-1 reads the first registered image (hierarchical image including the original image) stored in the HD 20-4 again (step S403), and repeats the processing from step S404 onward.

  In this case, since N = 1, the similarity between the first layer hierarchical image of the registered image and the first layer hierarchical image of the reference image is calculated (step S404), and the predetermined first layer hierarchical image is calculated. The reference value TH is read (step S405), and the calculated similarity is compared with the read first layer reference value TH (step S406). Here, the reference value TH of the first layer is determined according to the resolution of the layer.

  If the calculated similarity is equal to or higher than the first layer reference value TH (YES in step S406), the registered image read in step S403 is selected as a registered image candidate (step S407), and the number of selected registered image candidates is selected. M is set to M = M + 1 (step S408).

  Then, after confirming that the number M of registered image candidates has not reached 100 (NO in step S409), and confirming that other registered images are stored in the HD 20-4 (YES in step S410). ), The next registered image is read out from the HD 20-4 (step S403), and the processes in and after step S404 are repeated.

  If the calculated similarity is not equal to or higher than the reference value TH of the first layer (NO in step S406), after confirming that another registered image is stored in the HD 20-4 (step S410), from the HD 20-4 The next registered image is read (step S403), and the processing from step S404 is repeated.

  Thereby, the hierarchical images of the registered images are read from the HD 20-4 one after another, and the similarity between the first layer hierarchical image of the registered images and the first layer hierarchical image of the reference image is calculated and calculated. The similarity is compared with the first layer reference value TH, and a registered image whose calculated similarity is confirmed to be equal to or higher than the first layer reference value TH is selected as a registered image candidate.

  During this process, when all registered images are read from the HD 20-4 and there are no other registered images (NO in step S410), the control unit 20-1 has a hierarchical image with higher resolution. (YES in step S411), N = N-1 is set (step S412), and the process returns to step S403.

  As a result, the control unit 20-1 reads the first registered image (hierarchical image including the original image) stored in the HD 20-4 again (step S403), and repeats the processing from step S404 onward.

  In this case, since N = 0, the similarity between the 0th layer image J0 of the registered image and the 0th layer image I0 of the reference image is calculated (step S404), and the predetermined 0th layer is calculated. The layer reference value TH is read (step S405), and the calculated similarity is compared with the read zeroth layer reference value TH (step S406). Here, the reference value TH of the 0th layer is determined according to the resolution of the layer.

  If the calculated similarity is greater than or equal to the 0th layer reference value TH (YES in step 406), the registered image read in step S403 is selected as a registered image candidate (step S407), and the number of selected registered image candidates M is set to M = M + 1 (step S408).

  Then, after confirming that the number M of registered image candidates has not reached 100 (NO in step S409), and confirming that other registered images are stored in the HD 20-4 (YES in step S410). ), The next registered image is read out from the HD 20-4 (step S403), and the processes in and after step S404 are repeated.

  If the calculated similarity is not equal to or higher than the reference value TH of the 0th layer (NO in step S406), after confirming that another registered image is stored in the HD 20-4 (step S410), from the HD 20-4 The next registered image is read (step S403), and the processing from step S404 is repeated.

  Thereby, the hierarchical images of the registered images are read out from the HD 20-4 one after another, and the similarity between the 0th layer image of the registered image and the 0th layer image of the reference image is calculated and calculated. The similarity and the 0th layer reference value TH are compared, and a registered image for which the calculated similarity is confirmed to be greater than or equal to the 0th layer reference value TH is selected as a registered image candidate.

  If the number M of selected registered image candidates becomes 100 during the repetition of this process (YES in step S409), the control unit 20-1 ends the registered image candidate selection process and is shown in FIG. The process proceeds to step S207.

  In addition, during the repetition of this process, if there is no hierarchical image with a higher resolution before the number M of selected registered image candidates reaches 100 (NO in step S411), the control unit 20-1 is selected. It is checked whether there is a registered image candidate (step S413). If there is a selected registered image candidate (YES in step S413), the registered image candidate selection process is terminated, and the process proceeds to step S207 shown in FIG.

  On the other hand, if there is no selected registered image candidate (NO in step S413), that is, all registered images are read from the HD 20-4 for all layers of the first layer, the second layer, and the zeroth layer. Nevertheless, if the number of selected registered image candidates is 0, the collation result between the reference image and the registered image is set to be inconsistent (step S414), and the process is terminated. The collation result in this case is displayed on the liquid crystal display device 12.

  In the second embodiment, N = 2 is set in step S402, and the similar calculation between the registered image and the reference image is started from the second layer (lowermost layer) hierarchical image. However, in step S402, N = 1 and calculation of similarities between the registered image and the reference image may be started from the first layer image.

[Embodiment 3: Registered Image Candidate Selection Process 3]
FIG. 7 shows a flowchart of registered image candidate selection processing 3. In the third embodiment, the registered image candidate selection process 3 is combined with the registered image candidate selection processes 1 and 2 to obtain a final registered image candidate. In the third embodiment, the registered image candidate selected in the registered image candidate selection processes 1 and 2 is set as the first registered image candidate, and the registered image candidate selected in the registered image candidate selection process 3 is the first registered image candidate. 2 registered image candidates.

  In the registered image candidate selection process 3, the control unit 20-1 calculates a pupil distance as measurement information from the reference image obtained in step S205 (step S501), and the registered image to which the calculated pupil distance belongs. Is selected from the HD 20-4 (step S502). Then, the registered image of this inter-pupil distance group is selected as the second registered image candidate (step S503).

  Then, the control unit 20-1 combines the second registered image candidate selected in the registered image candidate selection process 4 with the first registered image candidate selected in the registered image candidate selection processes 1 and 2, and finally A candidate registered image is obtained.

  For example, as a first example of the third embodiment, the registration image candidate selection process 1 or 2 is performed on the second registration image candidate selected in the registration image candidate selection process 3 to obtain the final registration image candidate. To get.

  For example, as a second example of the third embodiment, the first registered image candidate selected in the registered image candidate selection process 1 or 2 and the second registered image candidate selected in the registered image candidate selection process 3 To obtain a final registered image candidate.

  For example, as a third example of the third embodiment, the first registered image candidate selected in the registered image candidate selection process 1 or 2 and the second registered image candidate selected in the registered image candidate selection process 3 To obtain a final registered image candidate.

  In the third embodiment, the inter-pupil distance is acquired as the measurement information from the reference image. However, the face size, the pupil size, the face curvature, and the like are acquired, and the first measurement is performed based on the acquired measurement information. A registered image candidate may be selected.

[Embodiment 4: Registration Image Candidate Selection Process 4]
FIG. 8 shows a flowchart of registered image candidate selection processing 4. In the fourth embodiment, this registered image candidate selection process 4 is combined with the registered image candidate selection processes 1 and 2 to obtain a final registered image candidate. In the fourth embodiment, the registered image candidate selected in the registered image candidate selection processes 1 and 2 is set as the first registered image candidate, and the registered image candidate selected in the registered image candidate selection process 4 is the first registered image candidate. 3 registered image candidates.

  In this registered image candidate selection process 4, the control unit 20-1 sets the original image (128 × 128 pixel image) of the reference image obtained in step S205 as a high-resolution image of the uppermost layer (0th layer). An image of 64 × 64 pixels (first low resolution image) is created in the lower layer (first layer) of the high resolution image, and 32 × 32 is formed in the lower layer (second layer) of the first low resolution image. A pixel image (second low-resolution image) is created (see step S601, FIG. 9B).

  Then, the control unit 20-1 reads the first registered image (hierarchical image including the original image) stored in the HD 20-4 (step S602). That is, for the first registered image, a high-resolution image (128 × 128 pixel image) located in the uppermost layer (0th layer) and a first layer located in the lower layer (first layer) of this high-resolution image. Low resolution image (64 × 64 pixel image) and a second low resolution image (32 × 32 pixel image) located in the lower layer (second layer) of the first low resolution image. (See FIG. 9A).

  After obtaining the hierarchical images I0, I1, I2 of the reference image and the hierarchical images J0, J1, J2 of the registered image, the control unit 20-1 reads a predetermined reference point, and searches for the corresponding region. The feature area for this is set in the hierarchical images J0, J1, J2 of the registered images (step S603).

  In this example, a reference point P1 predetermined as a point for defining a feature region is read from a reference point plate PL as shown in FIG. 10A set for the original image of the registered image, and this reference A region of 32 × 32 pixels centered on the point P1 is set as a feature region S0 in the hierarchical image J0 of the registered image. In this example, the reference point P1 is determined as a point where the right eye enters as a characteristic portion in the zone of the characteristic region S0 in the hierarchical image J0.

  Similarly, a 32 × 32 pixel area centered on a point corresponding to the reference point P1 is set as a feature area S1 for the hierarchical image J1 of the registered image. A region of 32 × 32 pixels centering on a point corresponding to the reference point P1 is set as the feature region S2. Since the hierarchical image J2 is an image of 32 × 32 pixels, the entire area of the hierarchical image J2 is set as the feature area S2.

  Then, the control unit 20-1 sets N = 2 (step S604), and searches for the corresponding area of the N-th layer hierarchical image of the reference image corresponding to the feature area of the N-th layer hierarchical image of the registered image ( Step S605). In this case, since N = 2, the corresponding area S2 'of the second layer hierarchical image I2 of the reference image corresponding to the feature area S2 of the second layer hierarchical image J2 of the registered image is searched by POC matching.

  Then, the control unit 20-1 calculates the similarity between the feature region S2 in the second layer hierarchical image J2 of the searched registered image and the corresponding region S2 ′ in the second layer hierarchical image I2 of the reference image ( In step S606), a predetermined reference value TH of the second layer is read (step S607), and the calculated similarity is compared with the read reference value TH of the second layer (step S608). Here, the reference value TH of the second layer is determined according to the resolution of the layer and the position of the feature region S2.

  If the calculated similarity is greater than or equal to the second layer reference value TH (YES in step S608), the registered image read in step S602 is selected as a registered image candidate (third registered image candidate) (step S609). The number M (initial value 0) of the selected registered image candidates is set to M = M + 1 (step S610).

  Then, after confirming that the number M of the selected registered image candidates has not reached 100 (NO in step S611), confirm that other registered images are stored in the HD 20-4 (step S611). (YES in S613), the next registered image is read from the HD 20-4 (step S602), and the processes in and after step S603 are repeated.

  If the calculated similarity is not equal to or higher than the reference value TH of the second layer (NO in step S608), it is confirmed that there is a hierarchical image with a higher resolution (YES in step S612), and N = N−1 ( Step S614) and the process returns to Step S605. In this case, since N = 1, the control unit 20-1 reflects the search result in the second layer hierarchy, and the reference corresponding to the feature region S1 of the first layer hierarchical image J1 of the registered image. A search for the corresponding region S1 ′ of the first layer image I1 of the image is performed by POC collation (step S605).

  Then, the control unit 20-1 calculates the similarity between the feature region S1 in the first layer hierarchical image J1 of the searched registered image and the corresponding region S1 ′ in the first layer hierarchical image I1 of the reference image ( In step S606, a predetermined reference value TH of the first layer is read (step S607), and the calculated similarity is compared with the read reference value TH of the first layer (step S608). Here, the reference value TH of the first layer is determined according to the resolution of the layer and the position of the feature region S1.

  If the calculated similarity is equal to or higher than the reference value TH of the first layer (YES in step S608), the registered image read in step S602 is selected as a registered image candidate (third registered image candidate) (step S609). The number M of the selected registered image candidates is set to M = M + 1 (step S610). Then, after confirming that the number M of the selected registered image candidates has not reached 100 (NO in step S611), confirm that other registered images are stored in the HD 20-4 (step S611). (YES in S613), the next registered image is read from the HD 20-4 (step S602), and the processes in and after step S603 are repeated.

  If the calculated similarity is not equal to or higher than the reference value TH of the first layer (NO in step S608), it is confirmed that there is a hierarchical image with a higher resolution (YES in step S612), and N = N−1 ( Step S614) and the process returns to Step S605. In this case, since N = 0, the control unit 20-1 reflects the search result in the first layer hierarchy, and the reference corresponding to the feature region S0 of the zeroth layer image J0 of the registered image. The corresponding area S0 ′ of the 0th layer image I0 of the image is searched by POC matching (step S605).

  Then, the control unit 20-1 calculates the similarity between the feature region S0 in the 0th layer image J0 of the searched registered image and the corresponding region S0 ′ in the 0th layer image I0 of the reference image ( In step S606), a predetermined reference value TH of the 0th layer is read (step S607), and the calculated similarity is compared with the read reference value TH of the 0th layer (step S608). Here, the reference value TH of the 0th layer is determined according to the resolution of the layer and the position of the feature region S0.

  If the calculated similarity is greater than or equal to the 0th layer reference value TH (YES in step S608), the registered image read in step S602 is selected as a registered image candidate (third registered image candidate) (step S609). The number M of the selected registered image candidates is set to M = M + 1 (step S610). Then, after confirming that the number M of the selected registered image candidates has not reached 100 (NO in step S611), confirm that other registered images are stored in the HD 20-4 (step S611). (YES in S613), the next registered image is read from the HD 20-4 (step S602), and the processes in and after step S603 are repeated.

  If the calculated similarity is not equal to or higher than the reference value TH of the 0th layer (NO in step S608), the control unit 20-1 checks whether there is a hierarchical image with higher resolution (step S612). In this case, since there is no hierarchical image with a resolution higher than that of the 0th layer (NO in step S612), after confirming that other registered images are stored in the HD 20-4 (YES in step S613), the HD 20- 4 reads the next registered image (step S602), and repeats the processing from step S603 onward.

  As a result, the hierarchical images of the registered images are read from the HD 20-4 one after another, and the similarity between the feature region of the hierarchical image of the registered image and the corresponding region of the hierarchical image of the reference image is calculated and calculated in order of increasing resolution. The registered image is compared with the reference value TH of that hierarchy, and a registered image for which the similarity is confirmed to be greater than or equal to the reference value TH in any hierarchy is selected as a third registered image candidate.

  When the number M of selected registered image candidates becomes 100 during the repetition of this process (YES in step S611), the control unit 20-1 ends the registered image candidate selection process. Also, during the repetition of this process, if there are no other registered images before the number M of selected registered image candidates reaches 100 (NO in step S613), the registered image candidate selection process is terminated. .

  Then, the control unit 20-1 combines the third registered image candidate selected in the registered image candidate selection process 4 and the first registered image candidate selected in the registered image candidate selection processes 1 and 2, and finally A candidate registered image is obtained.

  For example, as a first example of the fourth embodiment, the registered image candidate selection process 1 or 2 is performed on the third registered image candidate selected in the registered image candidate selection process 4 to obtain a final registered image candidate. To get.

  For example, as a second example of the fourth embodiment, the first registered image candidate selected in the registered image candidate selection process 1 or 2 and the third registered image candidate selected in the registered image candidate selection process 4 To obtain a final registered image candidate.

  For example, as a third example of the fourth embodiment, the first registered image candidate selected in the registered image candidate selection process 1 or 2 and the third registered image candidate selected in the registered image candidate selection process 4 To obtain a final registered image candidate.

  In the fourth embodiment, N = 2 is set in step S604, and the search for the corresponding region corresponding to the feature region (calculation of similarity) is started from the second layer (lowermost layer) hierarchical image. In step S604, N = 1 may be set, and a search for a corresponding region corresponding to the feature region and calculation of the similarity may be started from the first layer image.

  In the fourth embodiment described above, the feature areas S0, S1, and S2 are defined in the hierarchical images J0, J1, and J2 of the registered image, but the feature areas S0, S1 and S2 may be determined, and the corresponding areas S0 ′, S1 ′, and S2 ′ may be searched from the hierarchical images J0, J1, and J2 of the registered images.

  In the fourth embodiment, the registered image candidate selection process 4 is combined with the registered image candidate selection processes 1 and 2 to obtain the final registered image candidate. However, the registered image candidate selection process 3 And 4 may be combined with registered image candidate selection processes 1 and 2 to obtain a final registered image candidate.

  In the first and second embodiments, the registered image candidate is selected until M = 100. However, if the similarity is very high, the selection of the registered image candidate is terminated there. May be. For example, in the first embodiment (FIG. 4), when the similarity is greater than or equal to the predetermined value THmax that is much larger than the reference value TH in step S306, the selection of the registered image candidate is terminated there. . Similarly, in the fourth embodiment (FIG. 8), if the similarity is greater than or equal to the predetermined value THmax that is much larger than the reference value TH in step S608, the selection of the third registered image candidate is terminated there. May be. Thereby, speeding up can be achieved.

  In the first and second embodiments described above, the registered image candidates are selected until M = 100. However, all registered images are collated with each other for classification, and the similarity (score) is determined. May be selected as registered image candidates. For example, in the first embodiment (FIG. 4), the collation processing for classification is performed on all the registered images stored in the HD 20-4 to obtain the similarity, and the top M pieces having the highest similarity are obtained. A registered image is selected as a registered image candidate. Similarly in the fourth embodiment (FIG. 8), the matching process for classification is performed on all the registered images stored in the HD 20-4 to obtain the similarity, and the top M registered images having the highest similarity. May be selected as the third registered image candidate.

  In the above-described first to fourth embodiments, the similarity calculation method (for example, step S208 (FIG. 3), step S304 (FIG. 4), step S404 (FIG. 6), step S606 (FIG. 8), etc. Although the details of the similarity calculation method) have not been described, the following method can be considered as the similarity calculation method in this case.

[Method of calculating similarity for specific face data, specific hierarchy, and specific area]
Option 1. The average value of the correlation values for each corresponding point is set as the similarity.
Option 2. The number of corresponding points higher than the threshold T is set as the similarity.
⇒ It is difficult to be affected by disturbances (such as wearing items) in specific areas.
Option 3. The number of corresponding points higher than the threshold “T_high” for each area such as the entire top, bottom, left and right is set as the similarity. The threshold “T_high” may be changed for each area.
⇒ Collation that is not easily affected by sunglasses, glasses, whiskers, scratches, eye patch, etc. can be realized.
Option 4. The number of corresponding points higher than the threshold “T_high” for each area such as the entire top, bottom, left and right is set as the similarity. At this time, if there are a large number of corresponding points lower than the threshold “T_low” in any of the areas, it is determined as NG in the similarity determination process in the subsequent stage. The thresholds “T_high” and “T_low” may be changed for each area.
⇒ Other people's misidentification due to the same wearing can be prevented. For example, misidentification by glasses with thick frames is eliminated.

  For example, in the first embodiment (FIG. 4), the similarity is calculated in step S304, and the similarity is compared with the reference value in step S306. The comparison may be performed in one step. The same applies to other embodiments.

  In the first to fourth embodiments described above, the hierarchical image of the registered image is stored in the HD 20-4 as a registered image. However, only the original image of the registered image may be stored. When only the original image of the registered image is stored, after reading out the registered image (original image), the original image of the registered image that has been read out is used as a high-resolution image of the uppermost layer (0th layer). A first low resolution image is created in the lower layer (first layer) of the resolution image, and a second low resolution image is created in the lower layer (second layer) of the first low resolution image.

  FIG. 11 shows a flowchart in the case where a registered image (original image) is read and then a hierarchical image of the registered image is created in the registered image candidate selection process 1. In this case, the original image of the registered image is read in step S302, and step S315 for creating a hierarchical image of the registered image is provided after step S302.

  FIG. 12 shows a flowchart when a registered image (original image) is read out and a hierarchical image of registered images is created in the registered image candidate selection process 2. In this case, the original image of the registered image is read in step S403, and step S415 for creating a hierarchical image of the registered image is provided after step S403.

  FIG. 13 shows a flowchart when a registered image (original image) is read out and a hierarchical image of registered images is created in the registered image candidate selection process 4. In this case, the original image of the registered image is read in step S602, and step S615 for creating a hierarchical image of the registered image is provided after step S602.

  The collation device of the present invention is not limited to a system that manages entry / exit as a collation device that collates a reference image with a registered image, but a system that authenticates various objects (for example, image check such as printing, solder substrate check) Etc.).

  DESCRIPTION OF SYMBOLS 10,11 ... CCD camera, 10-1, 11-1 ... Lens, 12 ... Liquid crystal display (LCD), 20 ... Processing part, 20-1 ... Control part, 20-2 ... ROM, 20-3 ... RAM, 20-4: hard disk (HD), 20-5: frame memory (FM), 20-6: external connection unit (I / F), 20-7: Fourier transform unit (FFT), I0, I1, I2 ... see Hierarchical image of image, J0, J1, J2 ... Hierarchical image of registered image, S0, S1, S2 ... Feature region, S0 ', S1', S2 '... Corresponding region, PL ... Reference template, P1 ... Reference point.

Claims (7)

In a collation device that collates a reference image and a registered image,
For the reference image, the original image as the highest resolution high-resolution image, the high resolution image as the top layer, the high-resolution image and the same image area resolution only a resolution that different low low A hierarchical image creating means for creating a resolution hierarchical image;
Registered image storage means for storing a plurality of registered images having the same resolution as the reference image;
Registered image reading means for reading out the registered images stored in the registered image storage means one by one;
The same low-resolution hierarchy of the low-resolution hierarchical image of the reference image created by the hierarchical image creating means and the hierarchical image corresponding to the hierarchical image of the reference image of the registered image read by the registered image reading means Registration image candidate for calculating similarity with an image from an arbitrary low-resolution hierarchical image, and based on the calculated similarity , the registered image read by the registered image reading means is compared with the reference image And a first registered image candidate selection means for selecting as a verification apparatus. The verification device according to claim 1,
The first registered image candidate selection means includes:
When the calculated similarity satisfies a predetermined reference value, it is selected as a registered image candidate of a registered image read by the registered image reading unit. In the verification device according to claim 2,
The first registered image candidate selection means includes:
The collation apparatus characterized by using the reference value defined according to the resolution of the hierarchy as the reference value for each hierarchy of the hierarchy image. In the collation apparatus described in any one of Claims 1-3,
Measurement information acquisition means for acquiring predetermined measurement information from the reference image;
Second registered image candidate selection for selecting a registered image candidate to be compared with the reference image from among a plurality of registered images stored in the registered image storage unit based on the measurement information acquired by the measurement information acquisition unit Means and
The first registered image candidate selection means includes:
A collation apparatus that selects registered image candidates by combining registered image candidates selected by the second registered image candidate selecting means. The verification device according to claim 1,
It is predetermined for any one of the hierarchical image of the reference image created by the hierarchical image creating means and the hierarchical image corresponding to the hierarchical image of the reference image of the registered image read by the registered image reading means. A feature region setting means for setting a feature region;
Corresponding area search means for searching for a corresponding area corresponding to the feature area of one hierarchical image in which the characteristic area is set by the characteristic area setting means from the other hierarchical image;
Based on the similarity between the feature region set in one hierarchical image by the feature region setting unit and the corresponding region searched from the other hierarchical image by the corresponding region search unit, the feature image is read by the registered image reading unit. Second registered image candidate selection means for selecting the registered image as a registered image candidate to be compared with the reference image,
The first registered image candidate selection means includes:
A collation apparatus that selects registered image candidates by combining registered image candidates selected by the second registered image candidate selecting means. The verification device according to claim 1,
Measurement information acquisition means for acquiring predetermined measurement information from the reference image;
Second registered image candidate selection for selecting a registered image candidate to be compared with the reference image from among a plurality of registered images stored in the registered image storage unit based on the measurement information acquired by the measurement information acquisition unit Means,
It is predetermined for any one of the hierarchical image of the reference image created by the hierarchical image creating means and the hierarchical image corresponding to the hierarchical image of the reference image of the registered image read by the registered image reading means. A feature region setting means for setting a feature region;
Corresponding area search means for searching for a corresponding area corresponding to the feature area of one hierarchical image in which the characteristic area is set by the characteristic area setting means from the other hierarchical image;
Based on the similarity between the feature region set in one hierarchical image by the feature region setting unit and the corresponding region searched from the other hierarchical image by the corresponding region search unit, the feature image is read by the registered image reading unit. Third registered image candidate selection means for selecting the registered image as a registered image candidate to be compared with the reference image,
The first registered image candidate selection means includes:
A collation apparatus that selects a registered image candidate by combining the registered image candidate selected by the second registered image candidate selecting unit and the registered image candidate selected by the third registered image candidate selecting unit. In the collation apparatus described in any one of Claims 1-6,
The first registered image candidate selection means includes:
When the number of the selected registered image candidates is 0, the matching result between the reference image and the registered image is not matched.