JP2003203239A - Image data collating method, image data collating device, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce the processing quantity for collation, while maintaining and improving the probability of misidentifying an image from the same origin to be different or the probability of misidentifying an image from a different origin to be the same. <P>SOLUTION: An area having the maximum correlation respectively with a first area defined in a registered fingerprint image in a registered fingerprint image data storage part 20 is detected from a prescribed area in a collation fingerprint image in a collation fingerprint image data storage part 30, and similarity between both images based on the difference of mutual position relation between both areas is determined by a collation fingerprint determination part 44. In the case of similarity as a result, an area having the maximum correlation respectively with a second area defined in the registered fingerprint image is detected from a larger area than a prescribed area, and similarity between both images is determined by the collation fingerprint determination part 44. When the similarity is not found in either of two-time determinations, definition of arrangement on the first area and the second area is changed, and the former detection and determination between the two-time determinations are executed again following the changed arrangement definition. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for collating image data, and more particularly to a technique that contributes to a reduction in the amount of processing required for collating image data.

[0002]

2. Description of the Related Art One of the fields of application of the technique for collating image data is the field of fingerprint collation. Various techniques have conventionally been proposed as a method of collating fingerprint image data. Here, one of such techniques will be described.

In this matching method, pre-registered data representing a fingerprint image (hereinafter referred to as "registered fingerprint image data") and a subject who matches the registered fingerprint image data. The data representing the fingerprint image (hereinafter, referred to as “collation fingerprint image data”) is subjected to the processing shown in the following procedures (1) to (5) Matching is performed. (1) A plurality of rectangular areas A _i (where i = 1 to N, N is 2) in the registered fingerprint image A represented by the registered fingerprint image data.
The above integers) are defined and placed. (2) While scanning a rectangular area having the same shape as A _i in the collation fingerprint image B represented by the collation fingerprint image data,
The correlation coefficient between the pixel data in the rectangular area and the pixel data in A _i is calculated. (3) Of the rectangular areas in B, the one having the largest correlation coefficient calculated by the above-mentioned processing of (2), that is, the one showing the highest correlation with A _i is detected and the rectangle is detected. _Let it be area B _i . (4) The distribution of the rectangular areas A _i defined in A is compared with the distribution of the rectangular areas B _i detected from B, and the identity between A and B is evaluated based on the comparison result. (5) If the identity between A and B is denied in (4), A or B is rotated little by little within the range of a predetermined angle, and the processes of (1) to (4) are repeated. And the identity of B with.

[0004]

The evaluation method of the identity between A and B in the process (4) of the above-mentioned fingerprint image data collation method is performed by comparing the relative positions in the rectangular area distribution. If the deviation amount is smaller than a predetermined threshold value, both are evaluated as being the same, and if the deviation amount is larger, both are evaluated as not being the same. For this reason, if this threshold value is set small, the standard for determining that they are the same becomes high, so the probability of erroneously recognizing another person as the person (this probability is referred to as "others acceptance rate") is low. The probability of mistakenly identifying the person as the other person (this probability is referred to as "personal rejection rate") is high. On the other hand, if this threshold value is set to a large value, the false acceptance rate will decrease, but the false acceptance rate will increase.

As described above, the false rejection rate and the false acceptance rate are in a trade-off relationship, and in order to improve the false rejection rate and false acceptance rate, an increase in the amount of calculation is inevitable. In view of the above problems, in the collation of image data, the probability of misidentifying images of the same origin as not identical and the probability of misidentifying images of different origin as identical.
The problem to be solved by the present invention is to reduce the amount of processing for collation while improving.

[0006]

According to the present invention, image data representing a first image and image data representing a second image are collated and the first image and the second image are compared. The image data collation method and image data collation device for determining the degree of similarity with the image, and a program for causing a computer to perform the image data collation processing are premised.

Then, in the image data collating method according to one aspect of the present invention, the definition of the arrangement of the plurality of first areas and the plurality of second areas in the first image is acquired, and the first definition is obtained. For each of the regions, the first maximum correlation region having the maximum correlation with the first region is detected from a predetermined region in the second image, and the first detection is performed. The first determination of the degree of similarity is performed based on the difference between the positional relationship of the first maximum correlation region and the positional relationship between the first maximum correlation regions, and the degree of similarity by the first determination is the first predetermined degree. When satisfying, for each of the second region, the second maximum correlation region having the maximum correlation with the second region, the area of the predetermined area or more in the second image. Second detection to detect from area Performing a second determination of the degree of similarity based on the difference between the positional relationship between the second regions and the mutual positional relationship between the second maximum correlation regions, the degree of similarity by the first determination When does not satisfy the first predetermined degree, or when the degree of similarity by the second determination does not satisfy the second predetermined degree, the first region in the first image and Changing the definition of the placement for the second area,
The above-described problem is solved by performing the first detection and the first determination again according to the definition of the arrangement changed by the change.

Here, the first image and the second image are images representing a fingerprint, for example. In the above method, if the first image and the second image represent the same one, the second maximum correlation region having the maximum correlation with the partial region extracted from the first image The position on the image does not change even if the detection range of the maximum correlation region is narrowed down a little on the second image, so that the probability of misidentifying images of the same origin as not being the same is small. The present invention utilizes this fact.

That is, the positional relationship between the first regions whose arrangement is defined on the first screen and the positional relationship between the first maximum correlation regions detected by the first detection performed by narrowing the detection range. The first judgment to judge the degree of similarity of
Only when it is determined that the degree of similarity is high, the positional relationship between the second areas whose arrangement is defined on the first screen,
A second determination is performed to determine the degree of similarity with the positional relationship between the second maximum correlation regions detected by the second detection performed by expanding the detection range beyond the first detection.

When it is determined that the degree of similarity is low by the first determination performed based on the result of the first detection, even if the first detection having a narrow detection range is performed as described above, an image of the same origin is obtained. Since there is little influence on the probability of erroneously recognizing that they are not the same, it is not necessary to extend the detection range to make a determination, and the first image and the second image are different in the definition of the current area arrangement. It is possible to determine immediately. Therefore, according to the method described above, it is possible to reduce the processing amount for detection by the amount of narrowing the detection range in the first detection. Moreover, since the first detection is repeatedly performed while changing the definition of the area arrangement in the first image, the effect of reducing the processing amount increases according to the number of times of repetition.

In the second determination in the image data collating method according to the present invention described above, when the degree of similarity satisfies the second predetermined degree, the first image and the second image It can be determined that and represent the same thing.

By doing so, an increase in the probability of erroneously recognizing that images of different origins are the same, which may be caused by performing the first determination based on the first detection having a narrow detection range, is increased. Since it is canceled by performing the second determination based on the detection of, it is possible to reduce the processing amount for collation and to maintain / reduce the probability of misidentifying images of different origins as the same. The verification result can be presented.

Further, in the image data collating method according to the present invention described above, even if the definition of the arrangement of the first area and the second area in the first image is changed within a predetermined range, When the degree of similarity does not satisfy a predetermined degree in the first determination, it can be determined that the first image and the second image represent different images.

By doing so, it is possible to reduce the processing amount for the collation and to present the collation result of the image data in which the probability of erroneously recognizing that the images of the same origin are not the same is maintained. Further, by changing the definition in the image data matching method according to the present invention described above, the arrangement of the first region and the second region can be rotationally moved on the first image.

By doing so, although the first image and the second image represent the same image, the two images are different because a shift in the rotational direction occurs between the two images. The problem of misidentifying that there is is solved. In addition,
At this time, even when the degree of similarity in the second determination does not satisfy a predetermined degree, when the definition of the arrangement in the first image for the first region is changed. When the degree of similarity satisfies a predetermined degree in the first determination, it is possible to determine that the first image and the second image represent the same.

By doing so, the result of the determination that the degree of similarity satisfies the predetermined degree is obtained by the first determination performed again by rotating and moving the arrangement of the first region on the first image. If it is obtained again, the final matching result is that the ones shown in the first image and the second image are the same, so the second detection is the final matching result. Regardless of the reason, the first detection is always executed only once, and the first detection, which has a small detection processing amount due to the narrow detection range, is repeatedly executed. Therefore, the effect of reducing the processing amount is further enhanced. .

Further, in the above-described image data collating method according to the present invention, the definition of the arrangement of the first area and the arrangement of the second area may be the same. By sharing the definition of the arrangement in this way, the processing for the first and second detections can be made common except for the difference in the detection range, and the configurations for these detections are provided independently. There is no need.

Further, in the definition of the arrangement in the image data collating method according to the present invention described above, at least one arrangement in the first area is the same as any arrangement in the second area. Can be. By doing so, with respect to this common area, the positional relationship between the areas whose arrangement is defined on the first image and the mutual relationship between the maximum correlation areas detected from the second image are Since the first and second determinations can be made based on the difference, it is possible to partially share the first and second determinations, and it is necessary to separately provide a configuration for these determinations. Disappear.

Further, in the image data collating apparatus according to another aspect of the present invention, the storage for storing the definition of the arrangement of the plurality of first areas and the plurality of second areas in the first image is stored. A first maximum correlation region having a maximum correlation with the first region for each of the first regions, the first detection being performed from a predetermined region in the second image. One detection means, a first determination means for performing a first determination of the degree of similarity based on the difference between the positional relationship between the first regions and the positional relationship between the first maximum correlation regions, and When the degree of similarity by the first determination satisfies the first predetermined degree, for each of the second region, the second maximum correlation region having the maximum correlation with the second region, The predetermined in the second image Second detection means for performing a second detection for detecting from an area having a width larger than the area, and the similarity based on the difference between the positional relationship between the second areas and the positional relationship between the second maximum correlation areas. When the degree of similarity determined by the first determination does not satisfy the first predetermined degree, or the degree of similarity determined by the second determination A definition changing means for changing the definition of the arrangement of the first area and the second area in the first image when the second predetermined degree is not satisfied, and The first detection means and the first determination means solve the above-mentioned problems by performing the first detection and the first determination again according to the definition of the arrangement changed by the definition changing means. To do.

According to this structure, the same operation and effect as the above-described image data collating method according to the present invention can be obtained. Further, even in the case of a program for causing a computer to perform the process corresponding to the above-described image data matching method according to the present invention, by causing the computer to read and execute the program, the image data according to the present invention described above. The same action and effect as the collation method can be obtained, and the above-mentioned problems can be solved.

[0021]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings. Here, an embodiment in which the present invention is used for collation of fingerprint image data will be described. FIG. 1 is a diagram showing the configuration of a fingerprint image data collation device (hereinafter referred to as “this device”) for carrying out the present invention.

The fingerprint image data input unit 10 is for obtaining fingerprint image data, and is provided with, for example, an image scanner or an image sensor such as a CCD (Charge Coupled Device). Fingerprint image data input unit 10
FIG. 2 shows an example of a fingerprint image represented by the fingerprint image data acquired in step 2. Although the fingerprint image shown in the figure is binarized, the fingerprint image data may express a multi-gradation fingerprint image.

The fingerprint image data acquired by the fingerprint image data input unit 10 is registered and stored in the registered fingerprint image data storage unit 20. This registered fingerprint image data storage unit 20
The fingerprint image data stored in is treated as the registered fingerprint image data described above. Collation fingerprint image data storage unit 3
In 0, the fingerprint image data of the subject, which is acquired by the fingerprint image data input unit 10 and whose fingerprint is collated by this apparatus, that is, the collated fingerprint image data described above is temporarily stored.

The fingerprint collation processing unit 40 performs a process of collating the registered fingerprint image data stored in the registered fingerprint image data storage unit 20 with the collated fingerprint image data stored in the collated fingerprint image data storage unit 30. Be done. The fingerprint collation processing unit 40 includes a rotation conversion unit 41 and a template placement management unit 4
2, a maximum correlation detection unit 43, and a collation determination processing unit 44.

The rotation conversion unit 41 rotates the registered fingerprint image represented by the registered fingerprint image data input to the fingerprint collation processing unit 40 according to a predetermined instruction. The template placement management unit 42 has a plurality of regions of a predetermined size at predetermined positions in the registered fingerprint image after being processed by the rotation conversion unit 41 (these are collectively referred to as “template”). To place. Data indicating the size and arrangement of these areas are shown in the area arrangement table 42-1.

The maximum correlation area detection unit 43 is registered by the template arrangement management unit 42 in an area of a predetermined size in the collation fingerprint image represented by the collation fingerprint image data input to the fingerprint collation processing unit 40. The position of the one having the highest correlation with each area arranged in the fingerprint image is detected. Data indicating the detection range in the collation fingerprint image for detecting this area is shown in the detection area coordinate table 43-1.

The collation determination processing unit 44 compares the distribution of the plurality of areas arranged in the registered fingerprint image by the template arrangement management unit 42 with the distribution of the plurality of areas detected by the maximum correlation area detection unit 43. The identity of the registered fingerprint image and the collation fingerprint image is determined based on the comparison result.

In the collation result display section 50, the fingerprint collation processing section 4
The result of the collation processing performed by the collation determination processing unit 44 of 0 is displayed. This device is configured as described above. Note that the present apparatus includes a computer having a standard configuration, that is, a CPU that controls each component by executing a control program, a ROM, a RAM, and a magnetic storage device, and the CPU controls each component. A storage unit that is used as a storage area for a control program or a work area when the CPU executes the control program or a storage area for various data, an input unit that acquires various data corresponding to a user operation, a display, etc. An image scanner, an image sensor, etc. as the fingerprint image data input unit 10 in a computer including an output unit that presents various data to the user to notify the user and an I / F unit that provides an interface function for connecting to a network. Can also be configured by connecting.

Next, the fingerprint collation processing performed by this apparatus will be described. This fingerprint collation process is a process performed by the fingerprint collation processing unit 40 in FIG. 1, and the collation fingerprint image data acquired by the fingerprint image data input unit 10 and the registration fingerprint image data storage unit 20 previously stored. This is a process of making a collation determination with fingerprint image data.

The contents of this fingerprint collation process will be described below with reference to the flow chart shown in FIG. First, S1
In 01, the registered fingerprint image data is stored in the registered fingerprint image data storage unit 20
The collation fingerprint image data is read respectively. The fingerprint collation performed by this apparatus is premised on so-called one-to-one authentication, and the registered fingerprint image data that is collated with the collated fingerprint image data before execution of this fingerprint collation processing is, for example, It is assumed that it is uniquely specified in advance by the recognition code and password entered by the subject.

In step S102, -θmMAX is substituted for the parameter θm for setting the image shift correction angle, and the collation flag is set to "0". Here, the image misalignment correction angle is erroneously determined to be not the same even if the collation fingerprint image and the registered fingerprint image are obtained by the same subject, due to the misalignment in the rotation direction that may occur between them. In order to prevent this, it refers to the rotation angle in the rotation conversion performed on the registered fingerprint image before the collation determination. If this image shift correction angle is set to θm> 0, clockwise rotation conversion is performed, and if it is set to θm <0, counterclockwise rotation conversion is performed. Further, the constant θmMAX is a value indicating the maximum angle to be corrected by this rotation conversion, and θm is set to a value in the range of ± θmMAX.

The collation flag will be described later. In S103, the rotation conversion unit 41 rotationally converts the registered fingerprint image data read in the processing of S101 by the angle θm. Hereinafter, S104 to S11
By executing the process up to 4, the first matching process is performed between the registered fingerprint image data and the matching fingerprint image data.

At S104, the reference rectangular area K is registered on the registered fingerprint image represented by the registered fingerprint image data after the processing of the previous step is performed, and the template layout management unit is based on the definition of the area layout definition table 42-1. It is arranged by 42. In S105, the scanning range when the region related to the reference rectangular region K is detected from the collation fingerprint image represented by the collation fingerprint image data read in the process of S101 is the maximum detection region coordinates by the correlation region detection unit 43. Obtained from the table 43-1.

In S106, the maximum correlation area detection unit 43 scans the collation fingerprint image within the scanning range acquired by the processing in the previous step, and it is a rectangular area having the same shape and the same area as the reference rectangular area K. Then, the region KR having the maximum correlation coefficient with the reference rectangular region K is detected from within the scanning range.

In S107, the area arrangement definition table 42
The coordinate (X0, Y0) indicating the position on the registered fingerprint image of the reference rectangular area K defined as −1 is the collation determination processing unit 44.
The coordinates (XR0, YR0) that specify the position on the collation fingerprint image of the rectangular area KR detected by the processing of the previous step are sent to the maximum correlation area detecting unit 4.
3 and is sent to the collation determination processing unit 44 and stored therein.

The manner in which the above processing from S104 to S107 is performed will be described with reference to the drawings. FIG. 4 shows an example of a table included in the fingerprint collation processing unit 40. In the figure, the left side is an example of the area arrangement definition table 42-1 and the right side is the detection area coordinate table 43.
The example of -1 is shown.

The definition contents of the row "K" in the table example shown in FIG. 4 will be described with reference to FIG. First, based on the contents of the row "K" in the area arrangement definition table 42-1 as shown in FIG. 5A, a rectangle having a width W of "W0" and a height H of "H0" is set as a reference rectangular area K. The area is defined, and as shown in FIG. 5 (B),
It is defined that the reference rectangular area K is arranged at the position of coordinates (X, Y) = (“X0”, “Y0”) in the registered fingerprint image A. As the values of X0 and Y0, for example, X-axis and Y-axis values indicating the center position of the rectangular registered fingerprint image A are set.

It should be noted that in each of the figures after FIG. 5B,
In both cases, the right direction is the increasing direction of the X axis, and the downward direction is the increasing direction of the Y axis. In the process of S104, the reference rectangular area K is arranged on the registered fingerprint image A in accordance with these definitions.

Next, from the contents of the line "K" in the detection area coordinate table 43-1 as shown in FIG.
Coordinates (X0, Y0) indicating the arrangement of the reference rectangular area K in the registered fingerprint image A are used as the temporary reference in the collation fingerprint image B, and “± XMainSearch” in the horizontal direction (X-axis direction) with the temporary reference as the center, Vertical direction (Y-axis direction)
It is defined that the collation fingerprint image B is scanned within the range of "± YMainSearch".

In the processing of S105 and S106, this definition is acquired, the matching fingerprint image B is scanned based on the definition, and the correlation coefficient calculation target area is an area that is sequentially extracted from the matching fingerprint image B in the scan. Calculation of a correlation coefficient between KC and the reference rectangular area K in the registered fingerprint image A,
And the rectangular area K having the highest correlation with the reference rectangular area K
Detection from the R correlation coefficient calculation target area KC is performed. Then, in the processing of S107, the coordinates (XR0, YR0) for specifying the position in the collation fingerprint image B of the rectangular area KR detected as shown in FIG.
Coordinates indicating the position of the reference rectangular area K on the registered fingerprint image (X
0, Y0).

Here, the calculation of the correlation coefficient performed in the process of S106 will be described. In addition, here
Calculation of the correlation coefficient between the rectangular area A and the rectangular area B will be described. First, pixels included in each of the rectangular area A and the rectangular area B are respectively denoted by A (i, j), B (m,
n). However, the total number of pixels included in each of the rectangular area A and the rectangular area B is equal. Further, signal intensities, which are multi-gradation values indicating light and shade for these pixels, are defined as X _ij and Y _mn , respectively.

When these signal intensities are generalized and expressed as Z _pq , the following equation is defined.

[0043]

[Equation 1]

In the above equation, N represents the total number of pixels included in the rectangular area. Further, in the above equation, Σ indicates that it is the sum total of all the pixels included in the rectangular area. That is, the above equation represents the average value of the signal intensities of the pixels included in the rectangular area.

Next, the following equation will be further defined.

[0046]

[Equation 2]

The above equation shows the root mean square value of the signal intensity for the pixels included in the rectangular area. here,
The correlation coefficient C _AB between the rectangular area A and the rectangular area B can be calculated by the following equation expressed using the definition of the above equation.

[0048]

[Equation 3]

The correlation coefficient between both regions is calculated using the above equation. In the calculation of the correlation coefficient according to the above equation, instead of using the signal intensities of all the pixels in the rectangular area, for example, only the pixels lined up in an arbitrary line in the rectangular area are used. Calculation, using only pixels included in a part of the rectangular area, or using only pixels selected by arbitrarily thinning out the rectangular area There is no problem if the collation accuracy of the fingerprint collation can be obtained. The use of such a calculation method is advantageous because the number of pixels for which the correlation coefficient is calculated is reduced and the calculation amount is reduced. Further, another method of calculating the correlation coefficient may be adopted.

Returning to the explanation of FIG. In S108, S103
Rectangular regions P1 to P4 are arranged by the template arrangement management unit 42 based on the definition of the area arrangement definition table 42-1 on the registered fingerprint image represented by the registered fingerprint image data after the processing of.

In S109, the maximum correlation area detection unit 43 detects the scanning range when the area related to the rectangular areas P1 to P4 is detected from the collation fingerprint image represented by the collation fingerprint image data read by the processing of S101. Is acquired from the detection area coordinate table 43-1.

In S110, the maximum correlation area detection unit 43 scans the collation fingerprint image within the scanning range acquired in the previous step, and the rectangular area Pi
A region PRi having the same shape and the same area as the region R i having the maximum correlation coefficient with the rectangular region Pi is detected from the scanning range.

In S111, the area arrangement definition table 42
The coordinates (Xi, Yi) indicating the position on the registered fingerprint image of the rectangular area Pi defined as -1 are sent to and stored in the collation determination processing unit 44, and the rectangle detected by the processing in the previous step. The coordinates (XRi, YRi) that specify the position of the region PRi on the collation fingerprint image have the maximum correlation region detection unit 4
3 and is sent to the collation determination processing unit 44 and stored therein.

In S112, the above S109 to S1
The collation determination processing unit 44 determines whether the processing up to 11 has been completed for all of i = 1 to 4. If the determination result is Yes, the processing proceeds to S113, and if No, the processing returns to S109. Then, the above-described processing is repeated.

The manner in which the above-described processing from S108 to S112 is performed will be described with reference to FIGS. "P1" and "P1" in the table example shown in FIG.
Note the rows "2", "P3", and "P4". i = 1
4 to 4, from the contents of each of these lines, as shown in FIG. 6A, the rectangular region Pi has a width of Wi and a height of Hi.
6 is defined, and as shown in FIG. 6B, the rectangular area Pi is defined to be arranged at the position of coordinates (Xi, Yi) in the registered fingerprint image A. In the process of S108, the rectangular area Pi is arranged on the registered fingerprint image A according to these definition contents.

Further, from the contents of each line of Pi in the detection area coordinate table 43-1, the coordinates (Xi, Yi) indicating the arrangement of the rectangular area Pi in the registered fingerprint image A are used as the temporary reference in the collation fingerprint image B, and ± XSubSearchi in the horizontal direction and ± in the vertical direction around the temporary reference
It is defined that the collation fingerprint image B is scanned within the range of YSubSearchi. However, here
XSubSearchi <XMainSearch and YSubSearchi <YMainSearch, and the scanning range for detecting the rectangular area Pi is the rectangular area K.
It is narrower than the scanning range for detecting R.

In the processing of S109 and S110, these definitions are acquired, and the collation fingerprint image B based on those definitions is acquired.
Scan, the calculation of the correlation coefficient between the correlation coefficient calculation target area PCi, which is an area sequentially extracted from the collation fingerprint image B in the scan, and the rectangular area Pi in the registered fingerprint image A, and the correlation coefficient calculation Target area PCi to rectangular area Pi
The rectangular area PRi having the highest correlation with is detected. Here, the calculation of the correlation coefficient performed in the process of S110 may be performed by a method similar to that performed in the process of S106 described above, for example.

Note that FIG. 6C shows the detection range of the correlation coefficient calculation target area PC1 in the collation fingerprint image B for i = 1, that is, the rectangular area P1.
In the subsequent process of S111, the coordinates (X) for specifying the position of the detected rectangular region PRi in the collation fingerprint image B are detected.
Ri, YRi) is acquired and stored together with the coordinates (Xi, Yi) indicating the position of the rectangular area Pi on the registered fingerprint image A.

Note that FIG. 6D shows a state in which the rectangular area PR1 having the highest correlation with the rectangular area P1 arranged in the registered fingerprint image A is detected from the collation fingerprint image B. The above-described processing from S109 to S111 is the rectangular areas P1 to P4 by the determination processing of S112.
By performing the processing for each of the above, the arrangement of the rectangular areas used for the first matching performed between the registered fingerprint image data and the matching fingerprint image data is completed. An example of completion of arrangement of these rectangular areas is shown in FIG.

Returning to the explanation of FIG. In S113, the matching determination processing unit 44 calculates the difference in relative distance between K-Pi and KR-PRi, which is the basis of the matching determination between the registered fingerprint image data and the matching fingerprint image data. In subsequent S114, it is determined in the first collation process whether or not the result of the collation determination between both data based on this calculation result is OK, that is, both data represent the same fingerprint. Whether or not there is a collation determination processing unit 4
It is determined by 4. And this judgment result is Yes.
If so, the process proceeds to S117, and if No, the process proceeds to S115.

The above-mentioned S113 and S114 will be further described. By the processing of S107 and S111 described above, the collation determination processing unit 44 stores the data of the coordinates indicating the arrangement of the rectangular areas of K, KR, Pi, and PRi as shown in the table of FIG. Has been done. The characters shown in parentheses in the figure are ignored here.

In S113, the relative distance between the reference rectangular area K and the rectangular area Pi, and the rectangular area KR and the rectangular area PR.
The difference Δi from the relative distance between i and i is all i (= 1, 2,
3 and 4) are calculated based on the following equation.

[0063]

[Equation 4]

Then, in S114, it is determined whether or not all the calculated Δi are within a predetermined value. Here, if all Δi are within the predetermined value, the collation determination is OK, that is, it is determined that the fingerprints represented by both data are the same, while if not, the collation determination NG, That is, it is determined that the fingerprints represented by both data are not the same. Note that the predetermined value used here is, for example, Δ based on fingerprint image data acquired from a plurality of people.
i is actually calculated, and a value with which a desired matching accuracy is obtained is obtained based on the distribution of the calculation results.

Instead of the processing of S113 and S114, for example, a figure formed with K and Pi as vertices and a figure formed with KR and PRi as vertices,
It is also possible to adopt various image matching determination methods such as determination based on the difference in shape or area.

The description of the processing shown in FIG. 3 will proceed. S115
Then, the above-mentioned parameter θm in the rotation conversion unit 41.
The value of the constant n is added to the current value of, and the addition result is substituted into θm again. Although the value of the constant n is arbitrary, it should be noted that the smaller the value, the more finely the correction of the image shift can be performed, but the more the processing amount of the entire fingerprint matching process increases.

In S116, the rotation conversion unit 41 determines whether or not the current value of the parameter θm is less than or equal to the value of the constant θmMAX described above. If the determination result is Yes, the process returns to S103 and the registered fingerprint image data is returned. The angle θm is rotationally converted with respect to, and the first matching process after S104 is repeated. Therefore, after this, S1
Rectangular areas P1 to P1 arranged on the registered fingerprint image in the processing of 08
P4 and the rectangular areas Q1 to Q4 arranged on the registered fingerprint images in the second matching process described later are arranged at positions relatively rotationally moved from the positions on these registered fingerprint images last time.

On the other hand, if the determination result of S116 is No,
The final matching result is NG, that is, the matching result that the fingerprints represented by the registered fingerprint image data and the matching fingerprint image data are not the same is confirmed and the matching result is displayed on the matching result display unit 50. The matching process ends.

By the way, when the result of the above-mentioned discrimination processing of S114 is Yes, that is, when the collation result by the first collation processing is OK, whether or not the collation flag is set to "1" in S117. If the determination result is Yes, the final verification result is OK, that is, the verification result that the registered fingerprint image data and the verification fingerprint image data are the same fingerprint is confirmed, and the verification result is displayed in the verification result display unit 50. And the fingerprint collation process is completed. On the other hand, if the determination result in S117 is No, S11
At 8, the second matching process is performed. Details of this second matching process will be described later.

At S119, whether the result of the collation determination by the second collation processing performed at S118 is OK or not is determined.
That is, the matching determination processing unit 44 determines whether or not it is determined in the second matching processing that the registered fingerprint image data and the matching fingerprint image data represent the same fingerprint. If this determination result is Yes, the final matching result is OK, that is, the matching result that the registered fingerprint image data and the matching fingerprint image data are the same fingerprint is confirmed, and the matching result is displayed on the matching result display unit 50. This is displayed, and this fingerprint collation processing ends.

On the other hand, if the determination result in S119 is No, the collation flag is set to "1" in S120, and then the process returns to S115 to repeat the above-described process. That is, the collation flag is a flag indicating that the collation by the second collation processing is performed and the collation result is NG.

The above-described processing is the fingerprint collation processing, and the fingerprint collation processing unit 40 performs this processing to collate the fingerprint image data by this apparatus.
Next, the second collation process performed by the fingerprint collation processing unit 40 as the process of S118 in the above-described fingerprint collation process will be described. FIG. 9 is a flowchart showing the processing contents of the second matching processing.

The contents of the second collation process are basically the same as the processes from S108 to S113 in the above-mentioned first collation process existing in the flowchart shown in FIG. In FIG. 9, first, S2
In 01, a rectangular area Q1 is formed on the registered fingerprint image represented by the registered fingerprint image data after the processing of S103 of FIG.
To Q4 are arranged by the template arrangement management unit 42 based on the definition of the area arrangement definition table 42-1.

In S202, the scanning range when the region related to the rectangular regions Q1 to Q4 is detected from the collation fingerprint image represented by the collation fingerprint image data read by the process of S101 in FIG. 3 is the maximum correlation region. Detector 43
Is acquired from the detection area coordinate table 43-1.

In S203, the maximum correlation area detection unit 43 scans the collation fingerprint image within the scanning range acquired by the processing in the previous step, and the rectangular area Qi is detected.
An area QRi having the same shape and the same area as, and having the maximum correlation coefficient with the rectangular area Qi is detected from within the scanning range.

In S204, the area layout definition table 42
The coordinates (OPXi, OPYi) indicating the position on the registered fingerprint image of the rectangular area Qi defined as -1 are sent to the collation determination processing unit 44 and stored therein, and the rectangle detected by the processing in the previous step. Coordinates (OPXRi, OPYRi) that specify the position of the region QRi on the collation fingerprint image are acquired by the maximum correlation region detection unit 43 and sent to the collation determination processing unit 44 and stored therein.

In S205, the above-mentioned S202 to S2 are executed.
The collation determination processing unit 44 determines whether or not the processing up to 06 is completed for all of i = 1 to 4. If the determination result is Yes, the processing proceeds to S206, and if No, the processing returns to S202. Then, the above-described processing is repeated.

In S206, KQ, which is the basis of the collation judgment between the registered fingerprint image data and the collation fingerprint image data,
The difference between the relative distances between i and between KR and QRi is calculated by the collation determination processing unit 44, after which the second collation processing ends, and the processing returns to FIG. 3.

The above processing is the second collation processing.
The manner in which this processing is performed will be described with reference to FIGS. 4, 8, 10, and 11. "Q1", "Q2", "Q3", and "Q4" in the table example shown in FIG.
Pay attention to the line. When i = 1 to 4, the contents of each of these lines define a rectangular area having a width OPWi and a height OPHi as a rectangular area Qi, as shown in FIG. ), The rectangular area Q
i is the coordinate (OPXi, OPY in the registered fingerprint image A
It is defined to be placed at the position i). S201
In the process (1), the rectangular area Qi is arranged on the registered fingerprint image A according to these definitions.

In the present embodiment, OPWi =
Wi and OPHi = Hi, that is, the size of the rectangular area Qi is set to the registered fingerprint image A in the first matching process described above.
Is the same as the rectangular area Pi arranged in
= Xi and OPYi = Yi, that is, the position of the rectangular area Qi on the registered fingerprint image A is the same as the rectangular area Pi. By doing so, the second collation process and the first collation process can be made common, so that the configuration of the fingerprint collation processing unit 40 can be simplified.

Further, from the contents of each row of Qi in the detection area coordinate table 43-1, the coordinates (X0, Y0) indicating the arrangement of the above-mentioned reference rectangular area K in the registered fingerprint image A.
Is a temporary reference in the verification fingerprint image B, and the verification fingerprint image B is within the range of ± OPXSearchi in the horizontal direction and ± OPYSearchi in the vertical direction with the temporary reference as the center.
It is defined to perform a scan on. However, here, OPXSearchi = XMainSearch
Further, OPYSearchi = YMainSearch, and the scanning range for detecting the rectangular area Qi is the rectangular area K.
Same as the scanning range for detecting R, that is, rectangular area P
It is made wider than the scanning range for detecting i.

In the processing of S202 and S203, these definitions are acquired, and the collation fingerprint image B based on those definitions is acquired.
Scan, the calculation of the correlation coefficient between the correlation coefficient calculation target area QCi, which is an area sequentially extracted from the collation fingerprint image B in the scan, and the rectangular area Qi in the registered fingerprint image A, and the correlation coefficient calculation Rectangular area Q in the target area QCi
The rectangular area QRi having the highest correlation with i is detected. Here, the calculation of the correlation coefficient performed in the process of S203 may be performed by a method similar to that performed in the process of S106 described above, for example.

In FIG. 11A, i = 1,
That is, the detection range of the correlation coefficient calculation target area QC1 in the collation fingerprint image B for the rectangular area Q1 is shown. In the subsequent processing of S204, the detected rectangular area QR
The coordinates (OPXRi, OPYRi) for specifying the position of i in the collation fingerprint image B are acquired, and the rectangular area Qi
Coordinate indicating the position on the registered fingerprint image A of (OPXi, OP
It is stored together with Yi). By this processing, the collation determination processing unit 44 stores the data of the coordinates indicating the arrangement of the rectangular regions of Qi and QRi as indicated by the characters in parentheses in the table of FIG. It is stored together with coordinate data indicating the arrangement of rectangular areas.

11 (B) and 11 (C), the rectangular area QR1 having the highest correlation with the rectangular area Q1 arranged in the registered fingerprint image A is detected from the collation fingerprint image B. Shows. Here, FIGS. 11B and 11C showing the position of the rectangular region QR1 on the collation fingerprint image B, and FIG. 6D showing the position of the rectangular region PR1 on the collation fingerprint image B. ) With.

Rectangular area QR shown in FIG. 11 (B)
The position on the collation fingerprint image B of No. 1 corresponds to the position on the collation fingerprint image B of the rectangular region PR1 shown in FIG. That is, in this case, rectangular areas P1 and Q of the same size arranged at the same position on the registered fingerprint image A
The position on the matching fingerprint image B of the rectangular area having the highest correlation with each of 1 is different in the range of the area for which the correlation coefficient is calculated between the first matching processing and the second matching processing. Nevertheless, it is a coincidence. When such a result is obtained, it can be said that the registered fingerprint image A and the collation fingerprint image B are the same fingerprint with considerable accuracy.

On the other hand, the position of the rectangular area QR1 shown in FIG. 11C on the collation fingerprint image B is the same as the position of the rectangular area PR1 shown in FIG. 6D on the collation fingerprint image B. It's very different. That is, in this case, the positions on the collation fingerprint image B of the rectangular regions having the highest correlations with the rectangular regions P1 and Q1 of the same size arranged at the same positions on the registered fingerprint image A are in correlation with each other. If the range of the area for which the number is to be calculated is made different between the first collation process and the second collation process, it will change significantly. When such a result occurs, the registered fingerprint image A and the collation fingerprint image B are often not the same fingerprint.

The above-described processing from S202 to S204 is performed for each of the rectangular areas Q1 to Q4 by the determination processing of S205, so that the second processing is performed between the registered fingerprint image data and the collation fingerprint image data. The arrangement of the rectangular area used for collation is completed. FIG. 12 shows an example of the arrangement completion of these rectangular areas.

Comparing an example of the arrangement completion of the rectangular area by the second collation processing shown in FIG. 12 with that by the first collation processing shown in FIG. 7, the rectangular area in FIG. 7A and FIG. 12A is compared. 7B, each arrangement of Pi and each arrangement of the rectangular area Qi are the same on the registered fingerprint image A, but have the maximum correlation with each area. It can be seen that PR1 and QR1 in each arrangement of the rectangular area PRi and each arrangement of the rectangular area Qi in FIG. 12 (B) do not match on the collation fingerprint image B.

After that, by the process of S206, the difference in relative distance between K-Pi and KR-PRi, which is the basis of the collation determination between the registered fingerprint image data and the collation fingerprint image data, is calculated. It is calculated by the collation determination processing unit 44, but this process may be performed by the same method as that performed in the process of S113 described above.

When the process of S206 is completed, the process returns to S118 of FIG. 3, and whether or not the result of the collation determination between both data based on this calculation result is OK, that is, both data are the same. The collation determination processing unit 44 determines whether or not it is determined in the first collation processing that a fingerprint is expressed. If the process of S206 is performed by the same method as the process of S113 described above, the process of S118 may be the same technique as the process of S114 described above. The predetermined value used as the reference for the collation determination at this time may be the same as that used in the process of S114, but may be a different value.

In the collation determination in the above-mentioned second collation processing, the arrangement of the reference rectangular area K and the rectangular area KR is the same as that in the first collation processing. In this way, the determination in the first collation process and the determination in the second collation process are performed based on the positional relationship with the common area as a reference, so that the collation determination processing unit 4
4 can be partially shared, which simplifies the configuration.

Next, the advantages of the present device as compared with the fingerprint verification according to the above-mentioned conventional technique will be described. In this device, first, the first matching process is performed. In this process, the search range for detecting the rectangular area PRi having the maximum correlation with the rectangular area Pi arranged in the registered fingerprint image from the collation fingerprint image is set narrower than that in the conventional technique.

Even if such a setting is made, if the fingerprints shown in both images are the same, the detection position of the rectangular area PRi is unlikely to change, so the false rejection rate is largely different from that in the prior art. There is no. However, when the fingerprints shown in both images are not the same, the position of the rectangular area detected in the search range narrowed by such a setting is different from that detected in the search range based on the conventional technique. Occurrence may occur. When this occurs, the false acceptance rate, that is, the probability of mistakenly making the collation NG falsely the collation OK will be increased. That is, in the first matching process, the reliability of the matching NG can be assured as the reliability of the conventional technique, but the reliability of the matching OK is lower than that of the conventional technique.

Therefore, in the present apparatus, when the first matching process is OK, when the rectangular area QRi having the maximum correlation with the rectangular area Qi arranged in the registered fingerprint image is detected from the matching fingerprint image. The second collation process is performed by making the search range of (1) wider than the first collation process. If the search range of the second matching process is set equal to or higher than that in the conventional technique, for example, the false acceptance rate in the second matching process can be expected to be equal to or higher than that in the conventional technique. Therefore, by performing the first and second collation processes in this manner, it is possible to maintain the same false rejection rate as that of the conventional technique and to expect a better false acceptance rate than that of the conventional technique. It will be possible.

Further, in the first collation process, the search range described above is narrower than that in the prior art, so that the processing amount required to complete the collation in the first collation process is smaller than that in the conventional technique. It is also clear that it will be reduced. Moreover, this first matching process is repeated while the registered fingerprint image is gradually rotated within the range of the predetermined correction angle until the matching result of this process becomes OK, so that the processing amount can be reduced. It increases in proportion to the number of repetitions. On the other hand, the second matching process is only executed once when the matching result of the first matching process is OK. Therefore, even if the search range of the second matching process is set to be slightly wider than that in the conventional technique, the increase in the processing amount for this purpose is greater than the reduction in the processing amount by the repeated first matching process. Since the number is much smaller, the processing amount when the first and second collation processes are overlapped and executed as described above is reduced as a whole.

From the above, in the fingerprint collation by this apparatus, the processing amount for collation is reduced and the increase in the probability of misidentifying the same fingerprint represented in each of two images as not being the same is suppressed. However, it is possible to reduce the probability that different fingerprints shown in each of the two images are mistakenly recognized as the same.

In the fingerprint collation processing shown in FIG. 3, S1
Due to the effect of the discrimination processing of 17, even if the collation result by the second collation processing is NG, the first and the first fingerprints are registered between the registered fingerprint image and the collated fingerprint image which are rotationally converted at different correction angles. If the verification result of the verification process is OK twice,
The verification result that the fingerprints shown in the registered fingerprint image and the verification fingerprint image are the same is confirmed. This is because if the fingerprints shown in the registered fingerprint image and the collation fingerprint image are the same, the collation fingerprints of the rectangular regions KR and PRi are rotated even if the arrangement of the rectangular regions K and Pi is rotationally moved on the registered fingerprint image. Similarly, the arrangement on the image only rotationally moves, but if the fingerprints shown in the registered fingerprint image and the collation fingerprint image are different, the arrangement of the rectangular regions KR and PRi on the collation fingerprint image. Is based on the idea that should change significantly. That is, if the fingerprints shown in the registered fingerprint image and the collation fingerprint image are different,
It can be considered that the verification result of the first verification process between the registered fingerprint image and the verification fingerprint image that have been subjected to the rotation conversion at different correction angles is OK twice even considerably. Based on such an idea, when the collation result of the first collation process is OK twice, it is decided that the fingerprints shown in the registered fingerprint image and the collation fingerprint image are the same. It is a thing.

As a result, in the fingerprint collation process shown in FIG. 3, the second collation process is always executed only once regardless of the collation result between the registered fingerprint image and the collation fingerprint image. The first matching process with a small processing amount is repeatedly executed. From this, it can be said that the fingerprint collation process has a high effect of reducing the processing amount.

A modification of the above-described embodiment of the present invention will be described below. First, as a first modification, in the second matching process, the search range on the matching fingerprint image of the rectangular area QRi having the maximum correlation with the rectangular area Qi arranged in the registered fingerprint image is OPXSearchi> XSu.
bSearchi and OPYSearchi> YSub
Searchi, that is, if the rectangular area PRi having the maximum correlation with the rectangular area Pi arranged in the registered fingerprint image in the first matching process is set wider than the search range on the fingerprint image, at least the first matching is performed. The effect of improving the reliability of the collation NG by the processing is obtained.

As a second modification, OPWi <W
i and OPHi <Hi, that is, the width and height of the rectangular area Qi arranged in the registered fingerprint image in the second matching processing is the width and height of the rectangular area Pi arranged in the registered fingerprint image in the second matching processing. You may make it smaller than each. Also by doing so, the effect of improving the false acceptance rate by the second collation processing is exhibited.

As a third modification, contrary to the second modification, OPWi> Wi and OPHi> Hi, that is, the width of the rectangular area Qi arranged in the registered fingerprint image in the second collation processing and The height may be larger than the width and height of the rectangular area Pi arranged in the registered fingerprint image in the second matching process. Also by doing so, the effect of improving the false acceptance rate by the second collation processing is exhibited.

As a fourth modification, (OPXi,
OPYi) ≠ (Xi, Yi), that is, the rectangular areas Pi and Qi arranged in the registered fingerprint image in each of the first and second matching processes may be arranged at different positions. Here, for example, the arrangement of the rectangular area Qi on the registered fingerprint image in the second matching processing is located farther from the reference rectangular area K than the arrangement of the rectangular area Pi on the registered fingerprint image in the first matching processing. By setting K-Qi
Since the difference between the relative distance between the two and the relative distance between KR and QRi becomes more prominent than that in the first matching processing, the effect of improving the false acceptance rate by the second matching processing is further improved.

As a fifth modification, XSubSearchi and YSubSe in the above-described embodiment.
The value of archi, that is, the value indicating the search range on the matching fingerprint image of the rectangular area PRi having the maximum correlation with the rectangular area Pi arranged in the registered fingerprint image in the first matching processing is XASubSearchi and YASubSea.
xSubSearchi <XA
SubSearchi and Y SubSearchi <Y
ASubSearchi, and OPXSearch
i ≧ XASubSearch i and OPYSearch
i ≧ YASubSearchi, that is, the search range on the matching fingerprint image of the rectangular area PRi in the first matching processing is narrower than that in the above-described embodiment, and the matching fingerprint of the rectangular area QRi in the second matching processing is set. The search range on the image may be equal to or larger than the search range on the matching fingerprint image of the rectangular area PRi in the first matching process in the above-described embodiment. By doing so, the search range of the rectangular region PRi in the first matching process is further narrowed, so that a further reduction effect of the processing amount for fingerprint matching in this device can be expected.

As a sixth modification, when the result of the collation by the first collating process is OK, the second collating process is performed plural times instead of once, and the collating process by these collating processes is performed. Only when all the results are OK, the result of fingerprint collation by this device may be finally set to OK. In this case, in the second matching process performed multiple times,
For example, the registered fingerprint image and the collation fingerprint image rotated at different angles are targeted for collation. By doing so, the effect of improving the false acceptance rate in the entire second collation processing is further enhanced.

As a seventh modification, the areas K, Pi and Qi arranged in the registered fingerprint image and the areas KR, PRi and QRi detected from the collation fingerprint image are all limited to rectangular areas. However, these shapes can be arbitrarily configured.

It is also possible to combine two or more of the above first to seventh modified examples, except for the combination of the second modified example and the third modified example. In order to configure the fingerprint collation processing performed by the fingerprint collation processing unit 40 in the present apparatus described above using the computer having the standard configuration as described above,
This can be realized by creating a control program for causing a computer to perform the fingerprint collation processing shown in FIGS. 3 and 9 described above, and causing the computer to read and execute the control program.

The present invention can also be implemented by a computer by recording such a control program on a computer-readable recording medium and causing the computer to read the program from the recording medium and execute the program. . FIG. 13 shows an example of a recording medium that allows the computer to read the recorded control program. As shown in the figure, as the recording medium, for example, a storage device 62 such as a ROM or a hard disk device provided as an internal or external accessory device in the computer 61, or a flexible disk, MO (magneto-optical disk), CD- A portable recording medium 63 such as a ROM or a DVD-ROM can be used. Further, the recording medium may be a storage device 66 that is connected to the computer 61 via the network 64 and that is included in the computer that functions as the program server 65. In this case, the transmission signal obtained by modulating the carrier wave with the data signal expressing the control program is transmitted from the program server 65 through the network 64 which is the transmission medium, and the computer 61 demodulates the received transmission signal. The control program can be executed by playing back the control program.

In addition, the present invention is not limited to the above-described embodiment, and various improvements and changes can be made.

[0109]

As described in detail above, the present invention is
A first maximum correlation region having a maximum correlation with each of the plurality of first regions defined in the first image is detected from a predetermined region in the second image, and the first region has a mutual positional relationship. The similarity between both images is determined based on the difference between the positional relationship between the first maximum correlation regions. Here, when it is determined that the two images are similar to each other, the second maximum correlation region having the maximum correlation with each of the plurality of second regions defined in the first image is determined in the second image. From the area having a size larger than the predetermined area, and the similarity between the two images is determined based on the difference between the positional relationship between the second areas and the positional relationship between the second maximum correlation areas. In either of these two determinations, when it is determined that the two images are not similar,
The definition of the arrangement for the first region and the second region in the first image is changed, and the first detection and the first determination are performed again according to the changed definition of the arrangement.

By doing so, the probability of erroneously recognizing that images of the same origin are not the same is maintained, and the processing amount for collation is reduced while suppressing an increase in the probability of erroneously recognizing images of different origin as the same. Produce an effect.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a fingerprint image data collation device for implementing the present invention.

FIG. 2 is a diagram showing an example of a fingerprint image represented by fingerprint image data.

FIG. 3 is a flowchart showing processing contents of fingerprint collation processing.

FIG. 4 is a diagram illustrating an example of a table included in a fingerprint collation processing unit.

FIG. 5 is a diagram showing how a rectangular region KR is detected.

FIG. 6 is a diagram showing how a rectangular region PRi is detected.

FIG. 7 is a diagram showing an example of how rectangular regions to be matched in the first matching process are arranged.

FIG. 8 is a diagram illustrating an example of data input to a matching process determination unit.

FIG. 9 is a flowchart showing the processing contents of second matching processing.

FIG. 10 is a diagram showing how rectangular areas Qi are arranged.

FIG. 11 is a diagram showing how a rectangular area QRi is detected.

FIG. 12 is a diagram showing an example of how the rectangular areas to be matched in the second matching process are arranged.

FIG. 13 is a diagram showing an example of a computer-readable recording medium in which a recorded program is read.

[Explanation of symbols]

10 Fingerprint image data input section 20 Registered fingerprint image data storage 30 Collation fingerprint image data storage unit 40 Fingerprint matching processing unit 41 Rotation converter 42 Template placement management section 42-1 Area Allocation Definition Table 43 Maximum Correlation Area Detection Unit 43-1 Detection area coordinate table 44 Collation determination processing unit 50 Verification result display 61 Computer 62, 66 storage device 63 Portable recording medium 64 network 65 Program server

Claims (10)

[Claims] 1. The image data representing the first image and the image data representing the second image are collated to determine the degree of similarity between the first image and the second image. An image data matching method for determining, wherein definition of arrangement in the first image for a plurality of first areas and a plurality of second areas is acquired, and for each of the first areas, the first area The first maximum correlation region having the maximum correlation with the region of, the first detection to detect from a predetermined region in the second image, the positional relationship between the first regions and the first maximum correlation The first determination of the degree of similarity is performed based on the difference between the positional relationships between the regions, and when the degree of similarity by the first determination satisfies a first predetermined degree, the second For each of the areas of
The second maximum correlation region having the maximum correlation with the second region is subjected to a second detection for detecting from a region having a width equal to or larger than the predetermined region in the second image, the second A second determination is made as to the degree of similarity based on the difference between the positional relationship between regions and the positional relationship between the second maximum correlation regions, and the degree of similarity according to the first determination is the first predetermined value. Of the first region and the second region in the first image when the degree of similarity is not satisfied, or when the degree of similarity by the second determination does not satisfy the second predetermined degree. 2. The image data matching method, wherein the definition of the arrangement is changed, and the first detection and the first determination are performed again according to the definition of the arrangement changed by the change. 2. The image data matching method according to claim 1, wherein the first image and the second image are images representing fingerprints. 3. In the second determination, when the degree of similarity satisfies the second predetermined degree, it is determined that the first image and the second image represent the same thing. The image data matching method according to claim 1, wherein 4. Even if the definition of the arrangement of the first area and the second area in the first image is changed within a predetermined range, the degree of similarity is predetermined in the first determination. 2. The image data collating method according to claim 1, wherein the first image and the second image are distinguished from each other when they do not satisfy the above condition. 5. The image data matching method according to claim 1, wherein the arrangement of the first region and the second region is rotationally moved on the first image by changing the definition. . 6. Even when the degree of similarity does not satisfy a predetermined degree in the second determination, when the definition of the arrangement of the first region in the first image is changed. In the first determination, when the degree of similarity satisfies a predetermined degree, it is determined that the first image and the second image represent the same thing. Item 5. The image data collating method according to Item 5. 7. The image data matching method according to claim 1, wherein the definition of the arrangement of the first area and the definition of the second area are the same. 8. In the definition of the arrangement, the arrangement of at least one of the first areas is the same as the arrangement of any of the second areas. Image data matching method described. 9. The image data representing the first image and the image data representing the second image are collated to determine the degree of similarity between the first image and the second image. An image data collating apparatus for judging, comprising: storage means for storing definition of arrangement in the first image for a plurality of first areas and a plurality of second areas; and for each of the first areas A first maximum correlation region having a maximum correlation with the first region, the first detection unit performing first detection for detecting a first maximum correlation region from a predetermined region in the second image, and the first region mutual And a first determination means for performing a first determination of the degree of similarity based on a difference between the positional relationship between the first maximum correlation region and the first maximum correlation region, and the degree of similarity determined by the first determination is first. The second area when a predetermined degree of For each,
Second detecting means for performing a second detection for detecting a second maximum correlation area having the maximum correlation with the second area from an area having a width equal to or larger than the predetermined area in the second image; A second determination means for performing a second determination on the degree of similarity based on a difference between the positional relationship between the second regions and the positional relationship between the second maximum correlation regions, and the first determination When the degree of similarity by the above does not satisfy the first predetermined degree, or when the degree of similarity by the second determination does not satisfy the second predetermined degree, the first in the first image Definition changing means for changing the definition of the arrangement for the one area and the second area, wherein the first detecting means and the first determining means are changed by the definition changing means. According to the definition of placement An image data collating apparatus, wherein the first detection and the first determination are performed again. 10. The image data representing the first image and the image data representing the second image are collated to determine the degree of similarity between the first image and the second image. A program for causing a computer to perform the image data matching process to determine, the process of acquiring the definition of the arrangement in the first image for a plurality of first regions and a plurality of second regions, A first detection process of detecting a first maximum correlation region having a maximum correlation with the first region from a predetermined region in the second image for each of the first regions; Processing for performing a first determination of the degree of similarity based on the difference between the positional relationship between the regions and the positional relationship between the first maximum correlation regions, and the degree of similarity based on the first determination is first. Meets a certain degree of To come, for each of said second region,
A second maximum correlation area having a maximum correlation with the second area is detected from an area having a width equal to or larger than the predetermined area in the second image; A process of performing a second determination of the degree of similarity based on the difference between the positional relationship between the two regions and the positional relationship between the second maximum correlation regions, and the degree of similarity according to the first determination is the When the first predetermined degree is not satisfied, or when the degree of similarity by the second determination does not satisfy the second predetermined degree, the first region and the first image in the first image A process of changing the definition of the arrangement for the second area, and a process of performing the first detection process and the first determination process again in accordance with the changed definition of the arrangement, Let it take place Program for.