JP2000357227A - Image data collation method, image data collation device and storage medium stored with image data collation program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To maintain a certain collation precision even in the case of collating image data which is not necessarily satisfactory as a collation object. SOLUTION: A definition part 3 defines a first area and a second area having a prescribed positional relation to this first area as first image data 1a. A first determination part 4 determines an area, which has a maximum collation with the first area, as second image data 1b. A redefinition part 5 redefines the second area on a first image on the basis of the positional relation on a second image of the first maximum collation area, and a second determination part 6 determines an area which has the maximum collation with the second area redefined on the second image. A discrimination part 7 discriminates the degree of similarity between both images on the basis of the difference between the mutual positional relation between the first area and the second redefined area and that between the first and second maximum collation areas or the difference between the positional relation between second areas and that between second maximum collation areas.

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 for maintaining a constant collation accuracy even when collating image data which is not always good as a collation target. Is suitable for, for example, collation of fingerprint image data.

[0002]

2. Description of the Related Art Fingerprint image data (hereinafter referred to as "registered fingerprint data") registered in a database or the like in advance.
There is a fingerprint collation device that collates the fingerprint image data of the subject acquired this time with an image sensor or the like (hereinafter, referred to as “verification fingerprint data”) to confirm the individual of the subject.

An example of a fingerprint collation method performed by a conventional fingerprint collation apparatus will be described with reference to a flowchart shown in FIG. First, S1
In step 001, the collation fingerprint data is converted into multi-tone image data (hereinafter, referred to as “shading”) using an image sensor or the like.
(Referred to as “multi-tone verification fingerprint data”).

In step S1002, the image data obtained by simply binarizing the multi-tone collation fingerprint data without performing any secondary processing such as interpolation is referred to as "binary collation fingerprint data". Get).

[0005] Thereafter, in S1003, based on the binarized collation fingerprint data, characteristic points such as branch points and end points of wrinkles constituting the fingerprint (these characteristic points are called characteristic figures or manushas). ) Is extracted, and position information indicating the positions of these characteristic figures on the image of the verification fingerprint data is obtained.

In step S1004, the identity of the registered fingerprint data and the collation fingerprint data is evaluated by comparing the positional information acquired in the previous step with the positional information of the characteristic figure on the image of the registered fingerprint data.

[0007] Some conventional fingerprint matching devices perform fingerprint matching according to the above-described procedure. The applicant of the present application has filed Japanese Patent Application No. 10-372205 prior to the present application, and the specification attached to the earlier application describes an example of a fingerprint collation method different from that described above. I have. This method can maintain a constant collation accuracy even with fingerprint image data that is not sufficiently clear. The procedure will be briefly described with reference to FIGS.

In the flow chart shown in FIG. 10, first, in S2001, a rectangular area A _I (i = 1 to 1) in a multi-tone image indicated by registered fingerprint data (this data is hereinafter referred to as “A data”). N, where N is an integer of 2 or more). In the example of A data shown in FIG. 11 (a), shows an example of defining the rectangular area A _i in the case of N = 5.

[0009] In S2002, the collation fingerprint data (this data hereinafter will be referred to as "B data") to scan a rectangular area of the rectangular region A _i the same shape and size and for the in multi-tone image shown, The correlation coefficient between the rectangular area and A _I is sequentially calculated. The calculation of the correlation coefficient uses the following calculation method.

The pixels included in each of the rectangular area A and the rectangular area B are assumed to be A (i, j) and B (m, n), respectively, and the total number of the pixels included in each of the rectangular area A and the rectangular area B is defined as Equal. The signal intensities, which are multi-tone values indicating shading of each pixel in both regions, are represented by X _ij and Y
_mn , and when these signal intensities are generalized and expressed as Z _pq , the following equation showing the average value of the signal intensities of the pixels included in the rectangular area

[0011]

(Equation 1)

And the following equation indicating the root mean square of the signal intensity for the pixels included in the rectangular area:

[0013]

(Equation 2)

[0014] (N is the total number of pixels included in the rectangular area, sigma, respectively showing the sum of all of the pixels included in the rectangular area) defined by the <Z>, the following indicated by <Z ^2> formula

[0015]

(Equation 3)

By calculating the value of the rectangular area A,
Correlation coefficient C with rectangular area B_ABIs calculated. S200
At 3, in the image indicated by the B data,
Rectangular area A _iThe rectangular area where the correlation coefficient with
And rectangular area B_iAnd

In step S2004, the image indicated by the A data
All defined rectangular areas A_iRectangular area B corresponding to_iTo
Judge whether or not it was acquired in the image indicated by the B data, and judge
If the result is Yes, the process proceeds to S2005; if No, S is performed.
Return to 2002 and get B _iIs performed.
In the example of the B data shown in FIG.
Rectangular area A_iRectangular areas B corresponding to the definition example of_i
Indicates a state in which is acquired.

[0018] In S2005, by comparing the distribution of the arrangement of the rectangular region B _i in the image indicated by the distribution and B data of the arrangement of the rectangular area A _i in the image represented by A data, A data (registered fingerprint data ) And B data (verification fingerprint data) are evaluated for identity.

The method described in the specification of the above-mentioned prior application performs fingerprint collation by the above-described procedure.

[0020]

FIGS. 12 (a) and 12 (b) are diagrams for explaining the problems of the prior art.
Shows examples of images of registered fingerprint data (A data) and collation fingerprint data (B data) of the same person.

The A data shown in FIG.
As described above, the acquisition of collation fingerprint data was
And the like, the matching fingerprint data is
And it is obtained with a large bias toward one side (right side in the figure).
It is assumed that In the method of the earlier application described above,
Rectangular area A shown in FIG. _i
Area where the correlation coefficient in B data is maximum
B_iAre specified, for example, as shown in FIG. this
In this case, as is clear from the figure, the fingerprint image of the same person
Despite the fact that
Shape area A_iDistribution in the image and in the image indicated by the B data
Rectangular area B_iAnd the distribution of the arrangement
U. Therefore, even if the method of the earlier application is performed,
It is possible to mistakenly determine that the crest image is not the same person
The probability of matching errors is generally about 0.1%.
This is one of the factors that greatly reduces the required fingerprint matching accuracy.
I was

In view of the above problems, it is an object of the present invention to maintain a certain collation accuracy even when collation of image data which is not necessarily a good collation target is performed.

[0023]

SUMMARY OF THE INVENTION According to the present invention, when the image data constituting the first image and the image data constituting the second image are collated to determine the degree of similarity between the two images, A first region and one or more second regions having a predetermined positional relationship between the first region and the first region are defined on the first image, and the first region and the second region are defined on the second image. A first maximum correlation area, which is an area having a maximum correlation with the first area, is determined, and the second area is determined based on a positional relationship of the first maximum correlation area on the second image. Redefining on the first image, determining, on the second image, a second maximum correlation area, which is an area having a maximum correlation with the redefined second area, A positional relationship between the second region redefined and the first maximum correlation region and the second maximum phase. The degree of the similarity based on the difference between the positional relationship between the second regions and the redefined positional relationship between the second regions and the positional relationship between the second maximum correlation regions. Thus, the above-described problem is solved.

FIG. 1 is a diagram showing the principle configuration of the present invention. In the figure, the first image data 1a and the second image data 1b are collated to determine the degree of similarity of the images constituting the two, and the result is output as the determination result 2.

The first image data 1a is, for example, image data registered in advance in a database or the like, and corresponds to, for example, registered fingerprint data when applied to the above-described example of the fingerprint matching device.

The second image data 1b is, for example, image data obtained using an image sensor or the like, and corresponds to, for example, collation fingerprint data when applied to the above-described fingerprint collation apparatus.

The first and second image data (1
The relationship between a, 1b) and the registered fingerprint data and the verification fingerprint data may be interchanged. The definition unit 3 includes, on a first image composed of the first image data 1a, a first area and one or more second areas having a predetermined positional relationship between the first area and the first area. Is defined. The definition of these regions is arbitrary, and if each region is not the same, for example, it may include a portion overlapping between the regions.

The first deciding unit 4 places the first image defined by the definition unit 3 on the second image composed of the second image data 1b.
Is determined as the first maximum correlation area which is the area having the maximum correlation with the area.

The redefinition unit 5 determines the second area defined by the definition unit 3 based on the positional relationship on the second image of the first maximum correlation area determined by the first determination unit 4. Is redefined on the image. The redefinition by the redefinition unit 5 is determined based on, for example, a difference between the positional relationship of the first maximum correlation area on the second image and the positional relationship of the first area on the first image.

The second determining section 6 is an area having the maximum correlation with the second area redefined by the redefining section 5 on the second image composed of the second image data 1b. The second maximum correlation area is determined.

The determination unit 7 determines the positional relationship between the first area defined by the definition unit 3 and the second area redefined by the redefinition unit 5 and the first area determined by the first determination unit 4. The difference between the positional relationship between the maximum correlation area and the second maximum correlation area determined by the second determination unit 6 or the positional relationship between the second areas redefined by the redefinition unit 5 and the second Determining the degree of similarity between the images of the first image data 1a and the second image data 1b based on the difference between the second maximum correlation areas and the positional relationship determined by the determination unit 6; It outputs the determination result 2.

According to the configuration shown in FIG. 1, when the determination section 7 determines the degree of similarity between the first image data 1a and the second image data 1b, the first area defined by the definition section 3 is determined. A positional relationship on the first image, and a positional relationship on the second image of a first maximum correlation area, which is an area having a maximum correlation with the first area, determined by the first determination unit 4; The determination can be made by reflecting the difference.
If this is applied to the above-described example of the fingerprint collation device, even when the image of the collation fingerprint data is largely deviated to one side with respect to the image of the registered fingerprint data, such an imbalance occurs between the data. The fingerprint collation can be performed on the assumption that the fingerprint exists. Redefinition unit 5 and second determination unit 6
Is provided to compensate in advance for the collation judgment performed by the judgment unit 7 on such inappropriate image data. With the above configuration, a decrease in the matching accuracy of the image matching is reduced, and a constant matching accuracy can be maintained.

In the redefinition of the second area, if the positional relation of the first maximum correlation area on the second image satisfies a predetermined relation, the original definition of the second area is used as it is. Accordingly, the time required for the process for redefining the second area is not required, and the time required for image matching can be reduced. When a plurality of regions are defined when defining the second region, the positional relationship of the first maximum correlation region on the second image satisfies a predetermined relationship in the redefinition of the second region. In this case, even if the original definition is used as it is for a part of the plurality of defined second regions,
Reduces the time required to redefine the region
The time required for image matching can be reduced.

[0034]

Embodiments of the present invention will be described below with reference to the drawings. Here, an embodiment of the present invention in a fingerprint matching device will be described.

FIG. 2 is a diagram showing a configuration of a fingerprint collating apparatus embodying the present invention. This device includes a CPU 11, a storage device 12, a RAM 13, an image reading device 14, a display unit 15, and an input unit 16, and is interconnected via a bus 17.

The CPU 11 uses the RAM 13 as a work area in accordance with a control program stored in the storage device 12, and controls the overall operation of the fingerprint collation device (hereinafter, referred to as "this device"). It is. The CPU 11 performs a fingerprint collation process described later by executing the control program.

The storage device 12 is constituted by a ROM, a hard disk device, or the like. The storage device 12 stores the above-described control program read out by the CPU 11 immediately after the power of the device is turned on, and also stores the first image shown in FIG. The registered fingerprint of the subject corresponding to the data 1a is stored in advance.

The RAM 13 is a work memory used by the CPU 11 when executing the above-described control program.
The image reading device 14 is a device for acquiring a verification fingerprint of a subject corresponding to the second image data 1b shown in FIG.
(Charge-coupled device).

The display unit 15 is a display device, such as a CRT or a liquid crystal, for displaying a collation judgment result corresponding to the judgment result 2 shown in FIG. Input unit 16
Is an input device such as a keyboard device for the user of the present apparatus to instruct the CPU 11 to obtain a verification fingerprint or to start a fingerprint verification process described later.

The configuration of the present apparatus is not much different from the configuration provided in a standard computer. Therefore, it is possible to realize the functions of the present apparatus with such a standard computer. For this purpose, for example, the fingerprint collation processing program is read from a storage medium storing a fingerprint collation processing program for causing a computer to perform processing equivalent to the fingerprint collation processing described later, using a reading device of the computer corresponding to the storage medium. Then, for example, it may be temporarily stored in a main memory, and then executed by a central processing unit provided in the computer.

FIG. 3 shows an example of a computer-readable storage medium for storing a fingerprint collation processing program. As shown in the figure, as the storage medium, for example, a storage device 22 such as a ROM or a hard disk device provided as an internal or external accessory device in the computer 21, a floppy (registered trademark) disk, an MO (magneto-optical disk) , A portable storage medium 23 such as a CD-ROM, a DVD-ROM, or a storage device 26 attached to a program server 25 which is a computer connected to the computer 21 via a network 24.

Next, control performed by CPU 11 shown in FIG. 2 will be described. FIG. 4 is a flowchart showing the details of fingerprint collation processing, which is particularly related to the present invention, in the control processing of the entire apparatus realized by the CPU 11 executing the above-described control program. 5 and 6
Is an example of a fingerprint image used for explaining the fingerprint collation processing.
5 and 6 are the same fingerprint images as those shown in FIG. 12, and (a-i) to (a-iv) indicate the storage device 12.
Shows the registered fingerprint data (A data) of the subject, which is stored in advance, and (bi) to (b-iv) show the verification fingerprint data (B data) of the subject obtained by the image reading device 14. I have. Each image in both figures is a multi-tone image.

Hereinafter, the fingerprint collation processing performed by the CPU 11 will be described with reference to FIGS. 4 and 5 and FIG.
explain. When an instruction to start the fingerprint collation processing is given to the input unit 16, first, in S101, as shown in (a-i) of FIG. 5, a main template mt, which is one of the rectangular areas, is defined in the A data image. The size of the main template mt may be arbitrarily set as long as required collation accuracy of fingerprint collation is obtained. The coordinates specifying the position of the A data of the main template mt defined on the image, that is, the coordinates of the center of gravity of the rectangular main template mt in this case are (x ₀ , y ₀ ).

Subsequently, in S102, (a-i) in FIG.
), The main plate m
Sub template s which is a plurality of rectangular areas different from t
Define t _i (i ≧ 1). And i = 4 in the example shown in (a-i) in FIG. 5, the sub-template st ₁ ~s a four rectangle around the main template mt
The t ₄ are defined in the positional relationship shown in FIG. The size of the sub-template st _i also be any as long as the collation precision of a fingerprint collation required is obtained. Here the sub-template defined st _i coordinates specifying the position on the image of the A data, here, the center of gravity coordinates of the sub-template st _i is rectangular and (x _i, y _i).

Next, in S103, a rectangular area having the same shape and size as the main template mt is set on the B data image, and the rectangular area is scanned on the image while the rectangular area and the main template mt are scanned. mt is sequentially calculated. As a method of calculating the correlation coefficient,
Here, the method used in the fingerprint collation method shown in the above-mentioned prior application is used as it is, but another calculation method of the correlation coefficient may be adopted.

In S104, as a result of sequentially calculating the correlation coefficient, the above-described rectangular area in which the correlation coefficient with the main template mt is maximized is determined as an area MT. Coordinates for specifying the position of the area B on the image of the B data, which are coordinates (x
₀ , y ₀ ), the coordinates of the area MT corresponding to (X ₀ , Y ₀ )
And

Subsequently, in S105, the area MT is
Whether or not the data belongs to an effective area E set on the image of the data, specifically, coordinates (X ₀ ,
It is determined whether or not (Y ₀ ) is included in the range _{Ex, y} indicating the effective area E. If the determination result is Yes, the process proceeds to S107, and if No, the process proceeds to S106.

Here, the effective area E is an image of B data.
In the above, the region MT and the sub template st_iWith
The maximum correlation coefficient between the sub-templates
st _iBetween a rectangular area of the same shape and size as
The relative arrangement relationship of the B data on the image is the A data
Template mt and sub template on the image of
To st_iIf you maintain the relative placement relationship with
Of the B data on the image where the area MT may exist
It is shown. In the present embodiment,
Range E shown_{x, y}Is set on the image of the B data.

[0049]

(Equation 4)

FIG. 5 (b-i) shows the main template mt on the A data image as shown in FIG. 5 (a-i).
This shows a case where the range _{Ex, y} of the effective area E is set on the image of the B data when the sub-template and the sub template st _i are defined. In the above equation, W _i and H _i represent the width (length in the x direction) and the height (length in the y direction) of the sub template st _i , respectively, and WIDTH and HEIGH.
T is the width (length in the x direction) and height (length in the y direction) of the image of each of the A data and the B data having the same size.
Is shown. In addition, x _i Search, y _i Sear
ch may have a rectangular area on the B data corresponding to the sub-template st _I on the A data shifted due to a variation in the operation of obtaining the A data and the B data in the fingerprint image of the same person. And a value corresponding to about 4 to 6 pixels in pixel units of the image data is used, for example. Max [p, q] is a function for obtaining the larger value of p and q (if equal, either of them).

The above equation will be further explained. FIG. 7 shows the effective area E
It is a figure explaining the setting technique of. This figure explains α _x and β _x in the above equation, and shows only the upper part of the A data. Here, in order to simplify the explanation, the sub-templates st ₃ and st ₄ are not considered.

Referring to the above equation and FIG. 7, α _x = L
₁ , it can be easily understood that β _x = WIDTH−L ₂ . By setting the range of the effective area E in the x direction by the values of α _x and β _x obtained by the above formula, when the sub-templates st ₁ and st ₂ are defined so as to extend in the x direction, Is the range of the effective area E, that is, the area M when the above-described positional relationship is maintained.
The range in which T can exist is reduced, and the sub-template st ₁
And the range of the effective region E is widened if the st ₂ was defined to be closely spaced in the x-direction.

Returning to the description of FIG. In the above-described S105, when it is determined that the area MT does not belong to the effective area E, that is, as in the relationship between (a-ii) and (b-ii) in FIG. If the region MT on the image of the B data corresponding to the template mt is gone out of the range E _{x, y} of the effective region E, in S106, by correcting the position of the sub-template st _i a
Rearrange the data on the image. Sub template st _i
Are performed according to the procedure shown in FIG.

FIG. 8 is a flowchart showing the contents of the sub-template relocation processing. This process is a process of rearranging the position specified coordinates (x _i, y _i) the sub-template st _i that is located in are identified by the coordinates (xr _i, yr _i).

Referring to FIG. First, it is determined whether or not the area MT is located within the range of the effective area E in the x direction (S201 and S203).

As a result of the determination, x for specifying the position of the area MT
Coordinate X₀Represents the range of the effective area E in the x direction._xthan
If smaller, sub template st after rearrangement_iArrangement
Xr is the x coordinate that specifies the position_i= X_i+ (Α_x-X₀)
Then (S202), the process proceeds to S206. Also, X₀Has
Β indicating the range of the effective area E in the x direction_xIf it is larger than
Sub template st after rearrangement_iThe location of the
Xr_i= X _i− (X₀−β_x) And (S20
4), proceed to S206.

Here, if neither of the above cases applies, that is, if the value of X ₀ is greater than or equal to α _{x and} less than or equal to β _x , xr _i = x _i , that is, the value as it is (S205), proceed to S206.

Subsequently, the area MT falls within the range of the effective area E.
Is determined in the same manner in the y direction (S
206 and S208). Processing to be performed based on the result of this determination
The processing is the same as the processing performed in the x direction described above.
The y coordinate Y specifying the position of the area MT₀Is effective area E
Α indicating the range in the y direction of_yIf less than, after relocation
Sub template st_iY-coordinate that specifies the layout position of
Is yr_i= Y_i+ (Α_y-Y₀) (S207), and Y
₀Represents the range of the effective area E in the y direction._yLarger than
Then y r_i= Y_i− (Y₀−β _y(S20)
9). Also not applicable to either of these
Case, ie, Y₀Is α_yAnd β_yLess than
Then yr_i= Y_i, That is, the raw value
(S210).

According to the above procedure, all sub-templates st _i defined on the A data image are rearranged. (A-iii) of FIG. 6 shows a case where the area MT on the image of the B data is out of the range _{Ex, y} of the effective area E as shown in (b-ii) of FIG. 5 (a-i)
Sub-template st ₁ had been defined as shown in ~
It shows the result of performing relocation st _4. 6 (b-iii) shows the area MT corresponding to the main template mt in (a-iii) of FIG. 6. Since the position of the main template mt has not been moved, its position is not changed. Does not move from the position in (b-ii) of FIG.

Returning to the description of FIG. S105 of the determination process results Yes, i.e., when was unnecessary relocation of the sub-template st _i, or after performing the rearrangement of the sub-template st _i by the processing of S106, in S107, the sub-template st _i A rectangular area having the same shape and size is set on the image of the B data, and the rectangular area is scanned on the image, and the correlation coefficient between the rectangular area and the sub template st _i is sequentially calculated. . Here, the method of calculating the correlation coefficient is S
Although the method used in step 103 is used, another method of calculating the correlation coefficient may be employed.

Then, in S108, as a result of sequentially calculating the correlation coefficient, the above-described rectangular area having the maximum correlation coefficient with the sub-template st _i is obtained, and the area ST _{i is} calculated.
And

Subsequently, in S109, all sub-texts
Plate st_iRegion ST _iHas been detected
No, and if the determination result is Yes, proceed to S110.
No. On the other hand, if the determination result is No, the process returns to S107, and
Area ST_iIs repeated. In FIG.
(A-iv) and (b-iv) are shown in (a-iii) of FIG.
Sub template st rearranged as follows₁~ St_Four
, The corresponding area ST₁~ ST_FourGet
Of the main template mt and its corresponding
This is shown together with the area MT.

Then, in S110, the A data is displayed.
Main template mt and sub template
Plate st_iDistribution in the image shown in the B data
MT and ST at_iCompare with the distribution of the arrangement
A data (registered fingerprint data) and B data
Data (verification fingerprint data). This step
The identity assessment performed by the group is described as prior art.
Substantially the same as the processing performed in S2005 of FIG.
It is. In the evaluation of identity, the main template
Excluding the rate mt and the area MT from the evaluation target,
Plate st _iDistribution and area ST_iMinutes of placement
It is also possible to make an evaluation only by comparison with a cloth.

The fingerprint collation processing described above is executed by the CPU.
11 executes the fingerprint collation in this apparatus. In the embodiment of the present invention described above, the main template mt and the sub template s
Although the t _i , the area MT, and the area ST _i are all rectangular areas, the shapes of these areas are not limited to rectangles and may be arbitrary. The main template mt and region MT, and it is desirable size and shape of the sub-template st _i and the region ST _i is the same, even if some differences, the required collation precision of a fingerprint collation is Acceptable as long as it is obtained.

A main template m defined in advance
In the embodiment of the present invention described above, the relationship between the arrangement of t and the sub-templates st _i is such that the main template mt and the sub-templates st _i are in contact with each other as shown in FIG. had been defined by the relationship, it is also possible to carry out the definition of the positional relationship between the main template mt and sub-template st _i is that the definition of the positional relationship as partially overlap, or no away.

Further, in the sub-template rearrangement process shown in FIG. 8 which is executed when it is determined that the region MT does not belong to the valid region E, the sub-template s
Although the rearrangement was performed by moving the same distance for each of the t _i , the direction of the area MT deviated in the upper, lower, left, or right direction on the image of the B data of the effective area E was taken into consideration. The moving distance to the rearrangement position of each template st _i may be made different. For each of the sub-template st ₁ ~st ₄ of this fact has been defined as shown in (a-i) in FIG. 5, coordinates specifying the position after relocation (xr _i, yr _i) a case of obtaining the This will be specifically described by way of example.

First, regarding the x coordinate, X ₀ <α _x is obtained by the relationship between the x coordinate X ₀ specifying the position of the area MT and the above α _x and β _x indicating the range of the effective area E in the x direction. xr If _{i = x i + (α x} -X 0) [i = 2,3] xr i = x i + (α x -X 0) × u [i = 1,4] However, u is 0 ≦ u <1 constant _{α x ≦ X 0 ≦ β xr} i = x i β x in case of _x <xr _i = _{x i} in the case of _{X 0 - (X 0 -β x} ) [i = 1, _{4] xr i = x i -} (X 0 -β x) × u [i = 2,3] However, u finds the xr _i as a constant in the 0 ≦ u <1.

Next, regarding the y coordinate, Y ₀ <α _y by the relationship between the y coordinate Y ₀ specifying the position of the area MT and the aforementioned α _y and β _y indicating the range of the effective area E in the y direction. In the case of yr _i = y _i + (α _y -Y ₀ ) [i = 1,2] yr _i = y _i + (α _y -Y ₀ ) × v [i = 3,4] where v is 0 ≦ v <1 constant α _y ≦ _{Y 0} ≦ β in the case of _y is yr _i = _{y i} β y <in the case of _{Y 0 yr i = y i -} (Y 0 -β y) [i = 3 , 4] yr _i = y _i − (Y ₀ −β _y ) × v [i = 1,2] where v is a constant satisfying 0 ≦ v <1, and yr _i is obtained.

When sub-templates are rearranged in accordance with the above equation, especially if the rearrangement is performed with u = v = 0, even if the area MT does not belong to the effective area E, both sub-templates are relocated. Depending on the positional relationship, fingerprint matching can be performed with the originally defined layout without rearranging some sub-templates, in which case the time required for sub-template relocation processing is reduced As a result, the processing time of the fingerprint matching processing can be reduced.

[0070]

According to the present invention, as described in detail above,
In the collation determination of the image, since the collation determination reflecting the difference in the positional relationship on the image in the region having the largest correlation between the two image data to be collated is performed, the collation determination is not always good. Even if the image data is collated such that the image is acquired unevenly, the reduction in the collation accuracy of the image collation is reduced, and an effect that a constant collation accuracy can be maintained is obtained.

[Brief description of the drawings]

FIG. 1 is a diagram showing a principle configuration of the present invention.

FIG. 2 is a diagram illustrating a configuration of a fingerprint matching device that implements the present invention.

FIG. 3 is a diagram illustrating an example of a computer-readable storage medium that stores a fingerprint collation processing program.

FIG. 4 is a flowchart showing the details of a fingerprint collation process;

FIG. 5 is an example (No. 1) of a fingerprint image used for explaining the fingerprint collation device.

FIG. 6 is an example (No. 2) of a fingerprint image used for explaining the fingerprint collation device.

FIG. 7 is a diagram illustrating a method for setting an effective area E.

FIG. 8 is a flowchart showing processing contents of a sub-template relocation processing.

FIG. 9 is a flowchart showing a procedure of an example of a conventional fingerprint matching method.

FIG. 10 is a flowchart showing a procedure of a fingerprint collation method described in the specification of the prior application.

FIG. 11 is a diagram illustrating a fingerprint collation method described in the specification of the prior application.

FIG. 12 is a diagram illustrating a problem of the related art.

[Description of Signs] 1a First image data 1b Second image data 2 Determination result 3 Definition unit 4 First determination unit 5 Redefinition unit 6 Second determination unit 7 Determination unit 11 CPU 12 Storage device 13 RAM 14 Image reading Device 15 Display unit 16 Input unit 17 Bus 21 Computer 22 Storage device 23 Portable storage medium 24 Network 25 Program server 26 Storage device

Claims (10)

[Claims] 1. An image data forming a first image and a second image data
An image data collating method for collating image data composing an image of the image data and determining a degree of similarity between the two images, wherein a predetermined positional relationship between a first region and the first region is determined. One or more second regions having the first maximum correlation region, which is a region having a maximum correlation with the first region, on the second image. Determining, redefining the second area on the first image based on the positional relationship of the first maximum correlation area on the second image, redefining the second area on the second image Determining a second maximum correlation area, which is an area having a maximum correlation with the second area, and a positional relationship between the first area and the redefined second area; A difference in the positional relationship between the first maximum correlation area and the second maximum correlation area, or redetermination Image data verification method comprising determining, in that the degree of the similarity on the basis of the positional relationship between the second region mutual differences between the positional relationship between the second maximum correlation regions mutually, the. 2. A method of redefining the second region is defined by a positional relationship of the first maximum correlation region on the second image and a positional relationship of the first region on the first image. 2. The method according to claim 1, wherein the determination is made based on a difference between
Image data collation method described in 1. 3. The method according to claim 2, wherein the redefinition of the second area includes the first area.
2. The image data matching method according to claim 1, wherein when the positional relationship of the maximum correlation area on the second image satisfies a predetermined relationship, the original definition of the second area is used as it is. . 4. In the definition of the second area, a plurality of areas are defined, and in the redefinition of the second area, a positional relationship of the first maximum correlation area on the second image is predetermined. The image data matching method according to claim 1, wherein when the relationship is satisfied, the original definition is used as it is for a part of the plurality of defined second regions. 5. An image data forming a first image and a second image data
An image data collation method for collating image data composing an image of the first image and determining a degree of similarity between the two images, wherein a first region is defined on the first image, and the second image Determining a first maximum correlation area which is an area having a maximum correlation with the first area, and determining the degree of similarity with a positional relationship of the first area on the first image; The image data matching method, wherein the determination is made by reflecting a difference between the first maximum correlation area and the positional relationship on the second image. 6. The method according to claim 1, wherein the first and second images are fingerprint images.
Image data collation method described in 1. 7. An image processing system comprising: image data forming a first image;
An image data collation device that collates image data constituting the images of the two images to determine the degree of similarity between the two images, wherein a predetermined positional relationship between a first region and the first region is determined. Defining means for defining, on the first image, one or more second regions, and a first maximum, which is an area having a maximum correlation with the first area, on the second image. First determining means for determining a correlation area; and redefining means for redefining the second area on the first image based on a positional relationship of the first maximum correlation area on the second image. And second determining means for determining, on the second image, a second maximum correlation area which is an area having a maximum correlation with the redefined second area; and redefining the first area. The positional relationship between the second region and the first maximum correlation region and the second region. The similarity is determined based on a difference between a positional relationship with a maximum correlation area or a redefined difference between a positional relation between the second areas and the second maximum correlation area. An image data collating apparatus comprising: a determination unit configured to determine a degree of the image data. 8. An image processing system comprising: image data constituting a first image;
An image data collation device that collates image data constituting the image of the first and second images to determine the degree of similarity between the two images, wherein a definition unit that defines a first area on the first image, A first determining unit that determines a first maximum correlation area, which is an area having a maximum correlation with the first area, on the second image; and determining the degree of similarity with the first area of the first area. An image data collating device, comprising: a collation judging unit which makes a decision by reflecting a difference between a positional relationship on the image and a positional relationship of the first maximum correlation area on the second image. 9. The image data forming a first image and a second image data
A computer-readable storage medium storing an image data collation program for causing a computer to perform control to determine the degree of similarity between the two images by collating with image data constituting the image of the first image. And a control for defining, on the first image, an area and one or more second areas having a predetermined positional relationship with the first area; and, on the second image, Control for determining a first maximum correlation area that is an area having a maximum correlation with the first area; and a second area based on a positional relationship of the first maximum correlation area on the second image. Control on the first image, and control on the second image to determine a second maximum correlation area, which is an area having a maximum correlation with the redefined second area. The second area redefined as the first area Between the first region and the second region of maximum correlation, or the re-defined position of the second region. A storage medium storing an image data collation program for causing a computer to execute the control of determining the degree of similarity based on the difference between the second maximum correlation area and the positional relationship between the second maximum correlation areas. 10. An image data collation program for causing a computer to perform a control of collating image data constituting a first image with image data constituting a second image to determine the degree of similarity between the two images. A computer-readable storage medium storing: a control for defining a first area on the first image; and a maximum correlation with the first area on the second image. And determining the degree of similarity, the positional relationship of the first area on the first image, and the second image of the first maximum correlation area. And a storage medium storing an image data collation program for causing a computer to execute the control by making a determination by reflecting the difference from the above positional relationship.