JP3873167B2 - Image data collation device and image data collation processing program - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image data collating device for collating fingerprint images, for example. , And And an image data collation processing program.
[0002]
[Prior art]
For example, as a technical means for authenticating whether or not the person is the person, the fingerprint image registered in advance and the fingerprint image read for identification are collated to determine the similarity. A fingerprint image collating device that authenticates the user as a person when the degree of similarity is higher is considered.
[0003]
In such a fingerprint image collation apparatus, in order to determine the similarity between the registered fingerprint image (registration image) and the fingerprint image (collation image) read this time, first, a rectangular area is defined on the registration image. Then, an area having the same shape as this rectangular area is set on the collation image. Then, while changing the position of the set area on the collation image, the correlation coefficient of the image between the registered image and these areas on the collation image is obtained, and the rectangle on the collation image where the maximum correlation is obtained The similarity is determined based on the position of the region and the value of the correlation coefficient.
[0004]
[Problems to be solved by the invention]
However, in the fingerprint image collation device, when the fingerprint image is read, the angle of the subject's finger placed in the reading window is different each time. Therefore, when the registered fingerprint image and the collation fingerprint image are collated, While rotating the image, the correlation coefficient in which the rectangular area is set must be calculated for each rotated collation fingerprint image.
[0005]
Therefore, especially when the fingerprint reading angle in the verification fingerprint image is significantly different from the registered fingerprint image, the amount of calculation processing of the correlation coefficient with the registered fingerprint image while rotating the verification fingerprint image by a predetermined angle increases. However, it takes a long time until the determination result is obtained after the image of the verification fingerprint image is read.
[0006]
In addition, when reading a fingerprint image, the density of the fingerprint image is reduced each time due to the influence of external light that enters from the periphery of the subject's finger placed in the reading window and that is incident through the finger. If there is a large variation in shading between the registered fingerprint image and the verification fingerprint image, verification errors may occur even if the verification fingerprint image is the fingerprint image of the registered fingerprint image. There is a problem to become.
[0007]
The present invention has been made in view of the above-described problems. When collating fingerprint images, it is not rotated. Of Image data verification device that can capture fingerprints that are not suitable and easily perform fingerprint verification , And And an image data collation processing program.
[0008]
[Means for Solving the Problems]
That is, the image data collation apparatus according to the present invention is an image data collation apparatus for comparing and collating a fingerprint image registered in advance with a newly read fingerprint image, The correlation value at each rotation angle is calculated while rotating a predetermined ellipse pattern with respect to the fingerprint image data by a predetermined angle, and the rotation angle of the ellipse pattern at which the maximum correlation value is calculated is determined as the rotation angle of the fingerprint in the fingerprint image data. Tilt Inclination calculating means, inclination correcting means for matching the inclination of image data to be collated based on the inclination of fingerprint image data calculated by the inclination calculating means, and collation fingerprint image data in which the inclination of the fingerprint is adjusted by the inclination correcting means And image registration means for comparing the registered fingerprint image data.
[0012]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described below with reference to the drawings.
[0013]
FIG. 1 is a block diagram showing a configuration of an electronic circuit of a fingerprint collation apparatus 10 according to an embodiment of an image data collation apparatus of the present invention.
[0014]
The electronic circuit of the fingerprint collation apparatus 10 is configured by a computer and includes a control unit 11 including a CPU.
[0015]
The control unit (CPU) 11 is stored in a storage device by a control program stored in advance in the storage device 12 or a recording medium reading unit such as a floppy disk (R) drive FDD from an external recording medium 13 such as a floppy (R) disk FD. The fingerprint collation processing program read into the memory device 12 or the fingerprint collation processing program read into the storage device 12 from the external computer terminal (program server) via the communication network N via the communication network N is started, and the RAM 19 is The operation of each part of the circuit is controlled as a work memory.
[0016]
The storage device 12 is composed of a ROM of a semiconductor memory, a hard disk device HD, or the like, and stores in advance the control program that is read out by the control unit (CPU) 11 immediately after the verification device is turned on.
[0017]
The fingerprint collation processing control program stored in the storage device 12 includes fingerprint data collection processing (see FIG. 4), light intensity control processing (see FIG. 5), registration processing (see FIG. 6), and collation processing (FIG. 7) and tilt angle calculation processing (see FIGS. 8, 10, and 12).
[0018]
The RAM 19 is a work memory used by the control unit (CPU) 11 when executing the above-described control program.
[0019]
A control program for fingerprint collation processing stored in the storage device 12 is activated in response to an input signal from the input device 15, and the control unit (CPU) 11 includes the storage device 12 and communication control. In addition to the unit 14, the input device 15, and the RAM 19, the contact sensor 16 (16a to 16d), the image sensor 17, the light source 18 (18a, 18b), and the display device 20 are connected.
[0020]
The contact sensors 16 (16a to 16d) are arranged on the reading surface of the reading transparent plate 21 in the fingerprint reading unit (see FIG. 2), for example, at four corners of a fingerprint reading range H set in advance according to the fingerprint area of a general adult. It is provided as a transparent body and detects the contact of the finger when reading the fingerprint image. The contact with the finger is detected in all of the contact sensors 16 (16a to 16d), so that the fingerprint reading range H is correct. It is determined that a finger is placed at the position.
[0021]
The image pickup device 17 is provided in a fingerprint reading unit (see FIG. 2) and is composed of an image sensor such as a CCD (charge coupled device). The image pickup device 17 optically picks up a registered fingerprint image and a verification fingerprint image of a subject. The data is converted into an electrical signal corresponding to the gradation of the image and output. The imaging speed (shutter speed) of the imaging device 17 is variably set according to the amount of wraparound light entering from the periphery of the subject's fingerprint surface placed on the reading transparent plate 21. For example, when the amount of wraparound light is large, the shutter speed is set. The speed is fast and slow when it is low.
[0022]
The light source 18 (18a, 18b) is disposed below the reading transparent plate 21 in the fingerprint reading unit (see FIG. 2), and illuminates the fingerprint surface of the subject placed on the reading surface on the reading transparent plate 21. Thus, the amount of light emitted from the light source 18 is variably set according to the amount of wraparound light entering from the periphery of the fingerprint surface of the subject placed on the reading transparent plate 21. For example, when the amount of wraparound light is large, the light amount of illumination is set to increase, and when it is small, it is set to decrease.
[0023]
The display unit 20 is a display device such as a CRT or a liquid crystal display that displays a result of collation between the registered fingerprint image and the collation fingerprint image.
[0024]
The input device 15 is an input device such as a keyboard device for the user of the verification device to instruct the control unit (CPU) 11 to issue a fingerprint image collection instruction or start a fingerprint verification process described later.
[0025]
FIG. 2 is a diagram showing a configuration of a fingerprint reading unit in the fingerprint collation apparatus 10.
[0026]
This fingerprint reading unit is provided with a reading transparent plate 21 for placing a subject's finger in contact with the fingerprint surface, and in the four corners of the fingerprint reading range H on the reading surface of the reading transparent plate 21 are fingerprint images. Transparent body contact sensors 16a to 16d are provided for detecting the contact of the finger when reading.
[0027]
An imaging lens 22 is disposed on the lower side of the reading transparent plate 21 corresponding to the fingerprint reading range H, and the imaging lens 22 having the reading surface of the transparent plate 21 through the imaging lens 22 as a focal length. The image sensor 17 is disposed on the lower side.
[0028]
Light sources 18 a and 18 b are provided on both sides of the imaging lens 22 below the reading transparent plate 21 to illuminate the fingerprint surface of the subject's finger placed on the transparent plate 21.
[0029]
When reading the fingerprint image, it is determined that the finger is placed at the correct position in the fingerprint reading range H by detecting contact with the finger in all of the contact sensors 16a to 16d.
[0030]
FIG. 3 is a diagram showing a data memory secured in the RAM 19 of the fingerprint collation apparatus 10.
[0031]
The RAM 19 includes a first fingerprint image memory 19a, a second fingerprint image memory 19b, a light intensity setting memory 19c, a shutter speed setting memory 19d, an input fingerprint image memory 19e, a registered fingerprint image memory 19f, a verification fingerprint image memory 19g, and an inclination angle memory. 19h, a correlation value memory 19i and a work area 19j are secured, and an ellipse pattern memory 19k and a rotation angle / correlation value memory 19l are prepared.
[0032]
In the first fingerprint image memory 19a, fingerprint image data captured by the image sensor 17 without illumination by the light source 18 is stored as a first fingerprint image in the fingerprint data collection process (see FIG. 4). .
[0033]
In the second fingerprint image memory 19b, fingerprint image data captured by the image sensor 17 in the state of illumination by the light source 18 is stored as a second fingerprint image in the fingerprint data collection process (see FIG. 4). .
[0034]
The light amount setting memory 19c sets and stores the light amount for the light source 18 according to the wraparound light amount determined based on the first fingerprint image data stored in the first fingerprint image memory 19a.
[0035]
In the shutter speed setting memory 19d, the shutter speed for the image sensor 17 is set and stored according to the amount of sneak light determined based on the first fingerprint image data stored in the first fingerprint image memory 19a.
[0036]
The input fingerprint image memory 19e is imaged by the image sensor 17 according to the first fingerprint image data stored in the first fingerprint image memory 19a, the set light amount and the set shutter speed, and stored in the second fingerprint image memory 19b. The difference from the second fingerprint image data thus obtained is stored as the fingerprint image data GS of the subject.
[0037]
That is, the difference between the first fingerprint image data captured without illuminating with the light source 18 and the second fingerprint image data captured with illuminating with the light source 18 is collected as the fingerprint image data GS of the subject. Thus, the brightness portion around the fingerprint S or the influence of the external light transmitted through the fingerprint S is removed, and the fingerprint image GS in which the brightness of only the fingerprint portion S is obtained by illumination of the light source 18 is obtained. .
[0038]
When the fingerprint image data of the subject to be registered is collected in the registered fingerprint image memory 19f, the fingerprint image data GS stored in the input fingerprint image memory 19e is in a state in which the inclination of the rotation direction of the fingerprint is corrected. Is remembered.
[0039]
The fingerprint image data GS stored in the input fingerprint image memory 19e is stored in the collation fingerprint image memory 19g when the fingerprint image data of the subject to be compared with the registered fingerprint image is collected. Is stored in a corrected state.
[0040]
In the tilt angle memory 19h, for example, in the first tilt angle calculation process (see FIGS. 8 and 9), the fingerprint image data GS stored in the input fingerprint image memory 19e is image-recognized. A line connecting the maximum curvature positions of the pattern is used as a tilt axis J for fingerprint collection, and the tilt angle θ of this axis J is stored.
[0041]
In other words, the registered fingerprint image data stored in the registered fingerprint image memory 19f and the verification fingerprint image data stored in the verification fingerprint image memory 19g are both corrected for the inclination angle θ by the inclination axis J. Will be remembered.
[0042]
In the correlation value memory 19i, for example, correlation coefficient values calculated between image data of registered fingerprint image data and collation fingerprint image data are stored.
[0043]
In the work area 19j, various data input / output to / from the control unit (CPU) 11 in accordance with various control processes by the fingerprint collation apparatus 10 are temporarily stored and held as necessary.
[0044]
In the ellipse pattern memory 19k, the inclination angle θ of the fingerprint S in the fingerprint image data GS stored in the input fingerprint image memory 19e by the second inclination angle calculation processing (see FIGS. 10 and 11) is stored in a predetermined elliptic pattern P. The predetermined standard ellipse pattern P to be obtained by comparison correlation with is stored.
[0045]
In the rotation angle / correlation value memory 19l, in the second tilt angle calculation process, each rotation when the correlation coefficient value is calculated with the input fingerprint image data GS while rotating the standard ellipse pattern P by a certain angle. The angle and the correlation value are stored, and in the third tilt angle calculation process (see FIGS. 12 and 13), the center split axis X of the input fingerprint image data GS is rotated by a predetermined angle while the split axis X is centered. Each rotation angle and correlation value when the correlation coefficient values between the input fingerprint image data GS1: GS2 on both sides to be divided are calculated are stored.
[0046]
Next, the operation of the fingerprint collation device 10 configured as described above will be described.
[0047]
FIG. 4 is a flowchart showing fingerprint data collection processing by the fingerprint collation apparatus 10.
[0048]
When a fingerprint collection instruction operation is performed by the input device 15 when collecting a fingerprint image of a subject, the fingerprint image is provided on the reading transparent plate 21 of the fingerprint reading unit (see FIG. 2) corresponding to the fingerprint reading range H. Finger contact detection signals from the four contact sensors 16a to 16d are checked (step A1), and contact detection signals with the fingers are output from all the contact sensors 16a to 16d, that is, on the fingerprint reading range H It is determined whether or not the subject's fingerprint surface has been correctly placed (step A2).
[0049]
Here, even if one of the four contact sensors 16a to 16d does not output a contact detection signal with a finger, that is, the fingerprint surface of the subject is not correctly placed on the fingerprint reading range H. If it is determined, an error message indicating that the method of placing a finger on the fingerprint reading unit is bad is output to the display device 20, and an error is notified to the user (step A2 → A3).
[0050]
On the other hand, when contact detection signals with fingers are output from all the contact sensors 16a to 16d, that is, when it is determined that the subject's fingerprint surface is correctly placed on the fingerprint reading range H, illumination by the light source 18 is performed. In the state where is turned OFF, the fingerprint S of the finger corresponding to the fingerprint reading range H on the reading transparent plate 21 is imaged as the first fingerprint image data by the image sensor 17 and stored in the first fingerprint image memory 19 a in the RAM 19. Stored (step A2 → A4).
[0051]
Subsequently, with the illumination by the light source 18 turned on, the fingerprint S of the finger corresponding to the fingerprint reading range H on the reading transparent plate 21 is imaged as second fingerprint image data by the imaging device 17 and stored in the RAM 19. It is stored in the second fingerprint image memory 19b (step A5).
[0052]
Then, each pixel data of the first fingerprint image data without illumination stored in the first fingerprint image memory 19a and the second fingerprint image data with illumination stored in the second fingerprint image memory 19b. The fingerprint image data GS from which the difference for each is calculated and the brightness due to the wraparound of external light is removed is obtained and stored in the input fingerprint image memory 19e (step A6).
[0053]
FIG. 5 is a flowchart showing a light amount control process accompanying the fingerprint data collection process of the fingerprint collation apparatus 10.
[0054]
In this light quantity control process, the light quantity by the light source 18 and the imaging speed (shutter speed) by the image sensor 17 are optimally controlled in the fingerprint data collection process (see FIG. 4). First, in step A4, the light source 18 When the first fingerprint image data in the state where the illumination by is turned off is captured, the amount of sneak in ambient light is determined based on the first fingerprint image data (step B1).
[0055]
Then, based on the sneak amount of external light, the light amount by the light source 18 and the imaging speed (shutter speed) by the image sensor 17 are set and controlled (steps B2 and B3).
[0056]
For example, when the amount of sneak light due to external light is large, the light amount of the illumination by the light source 18 is set to increase, and when it is small, it is set to decrease. In the case where there are few, the setting is delayed so that the adverse effect of the external light on the fingerprint image is suppressed as much as possible. As a result, the periphery of the fingerprint S or the fingerprint S is transmitted to the input fingerprint image memory 19e. The brightness portion due to the influence of the surrounding light is removed, and fingerprint image data GS in which the brightness of only the fingerprint portion S is obtained by illumination of the light source 18 can be obtained.
[0057]
FIG. 6 is a flowchart showing fingerprint image registration processing by the fingerprint collation apparatus 10.
[0058]
When the fingerprint image data GS of the subject to be registered is collected and stored in the input fingerprint image memory 19e in the RAM 19 according to the fingerprint data collection process in FIG. 4 and the light amount control process in FIG. 5 (step C1), FIG. In accordance with the tilt angle calculation process in step 8, the tilt angle θ of the fingerprint S in the input fingerprint image data GS to be registered is calculated (step CE).
[0059]
Then, the fingerprint image data GS to be registered, which is stored in the input fingerprint image memory 19e by the inclination angle θ of the fingerprint S, is rotationally corrected, and the fingerprint image data GS without the inclination of the fingerprint S is registered as the registered fingerprint. It is stored and registered in the image memory 19f (step C2).
[0060]
FIG. 7 is a flowchart showing a fingerprint image collation process by the fingerprint collation apparatus 10.
[0061]
When the fingerprint image data GS of the subject to be collated is collected according to the fingerprint data collection process in FIG. 4 and the light quantity control process in FIG. 5 and stored in the input fingerprint image memory 19e in the RAM 19 (step D1). According to the tilt angle calculation process in FIG. 8, the tilt angle θ of the fingerprint S in the input fingerprint image data GS to be collated is calculated (step DE).
[0062]
Then, the fingerprint image data GS to be collated stored in the input fingerprint image memory 19e by the inclination angle θ of the fingerprint S is rotationally corrected, and the fingerprint image data GS without the inclination of the fingerprint S is converted into the collation fingerprint. After being stored in the image memory 19g, the verification fingerprint image data in which the inclination of the fingerprint S is corrected and the registered fingerprint image data in which the inclination of the fingerprint S is corrected and stored in the registered fingerprint image memory 19f as described above. The correlation coefficient value is calculated by comparison and stored in the correlation value memory 19i (step D2).
[0063]
Then, it is determined whether or not the correlation coefficient value between the registered fingerprint image data and the verification fingerprint image data is equal to or greater than a preset threshold value. If it is determined that a correlation value equal to or greater than the threshold value is obtained, the registered fingerprint image And the collation fingerprint are determined to match, and a determination message of identity matching is output from the display device 20, and if it is determined that the correlation value is less than the threshold value, the registered fingerprint and the collation fingerprint are determined to be inconsistent. An identity mismatch determination message is output from the display device 20 (step D3).
[0064]
(First tilt angle calculation process)
FIG. 8 is a flowchart showing a fingerprint image registration process and a first tilt angle calculation process accompanying the fingerprint image verification process by the fingerprint verification apparatus 10.
[0065]
FIG. 9 is a diagram showing the calculation principle of the fingerprint tilt angle θ accompanying the first tilt angle calculation process of the fingerprint collation apparatus 10, and FIG. 9A shows the reading state of the fingerprint S corresponding to the fingerprint reading range H. FIG. 6B is a diagram showing a storage state of the read fingerprint image data GS in the input fingerprint image memory 19e.
[0066]
That is, the fingerprint image data GS of the subject to be registered is collected in the fingerprint image registration process (see FIG. 6), and the fingerprint image data GS of the subject to be verified is collected in the fingerprint image matching process (see FIG. 7). When stored in the input fingerprint image memory 19e in the RAM 19, the process proceeds to the first inclination angle calculation process. First, the fingerprint S pattern in the fingerprint image data GS is image-recognized, and the maximum in each of the ridge lines of the plurality of fingerprints is detected. A position to be a curvature is obtained (step E1).
[0067]
Then, an axis J connecting the maximum curvature positions of the respective fingerprint ridge lines in the fingerprint image data GS is set (step E2), the inclination angle θ of the axis J is calculated, and the inclination of the fingerprint S in the input fingerprint image GS is calculated. The angle θ is obtained and stored in the tilt angle memory 19h (step E3).
[0068]
(Second tilt angle calculation process)
FIG. 10 is a flowchart showing a fingerprint image registration process and a second tilt angle calculation process accompanying the fingerprint image collation process by the fingerprint collation apparatus 10.
[0069]
FIG. 11 is a diagram showing the calculation principle of the fingerprint tilt angle θ associated with the second tilt angle calculation process of the fingerprint collation apparatus 10, and FIG. 11A shows a fingerprint image read corresponding to the fingerprint reading range H. The figure which shows the storage state in the input fingerprint image memory 19e of the data GS, the figure (B) is a figure which shows the standard ellipse pattern P from which the read fingerprint image data GS and a correlation coefficient are calculated.
[0070]
That is, the fingerprint image data GS of the subject to be registered is collected in the fingerprint image registration process (see FIG. 6), and the fingerprint image data GS of the subject to be verified is collected in the fingerprint image matching process (see FIG. 7). When stored in the input fingerprint image memory 19e in the RAM 19, the process proceeds to the second tilt angle calculation process, and first, the standard ellipse pattern P stored in the ellipse pattern memory 19k is read (step E11).
[0071]
Then, while the standard ellipse pattern P is rotated by a predetermined angle, the standard ellipse pattern P at each rotation position is correlated with the fingerprint image data GS collected and stored in the input fingerprint image memory 19e. The number is calculated, and the correlation coefficient value between each rotation angle of the elliptic pattern P and the fingerprint image data GS calculated corresponding to each rotation angle is sequentially stored in the rotation angle / correlation value memory 19l ( Steps E12 to E13).
[0072]
Then, for example, a correlation coefficient with the input fingerprint image data GS at each rotation angle when the elliptical pattern P is rotated by a predetermined angle in a range of ± 90 ° is calculated and stored in the rotation angle / correlation value memory 19l. When it is determined that the calculation of the correlation value over the entire range of the set rotation angle is completed (step E13), the input corresponding to each rotation angle of the elliptic pattern P stored in the rotation angle / correlation value memory 19l. Of the correlation values with the fingerprint image data GS, the rotation angle at which the maximum correlation value is obtained is obtained as the tilt angle θ of the fingerprint S in the input fingerprint image GS, and stored in the tilt angle memory 19h (step). E15).
[0073]
(Third tilt angle calculation process)
FIG. 12 is a flowchart showing a fingerprint image registration process and a third tilt angle calculation process associated with the fingerprint image collation process by the fingerprint collation apparatus 10.
[0074]
FIG. 13 is a diagram showing the calculation principle of the fingerprint tilt angle θ associated with the third tilt angle calculation process of the fingerprint collating apparatus 10, and FIG. 13A shows a fingerprint image read corresponding to the fingerprint reading range H. The figure which shows the storage state in the input fingerprint image memory 19e of the data GS, (B) and (C) in FIG. It is a figure which shows the correlation state in.
[0075]
That is, the fingerprint image data GS of the subject to be registered is collected in the fingerprint image registration process (see FIG. 6), and the fingerprint image data GS of the subject to be verified is collected in the fingerprint image matching process (see FIG. 7). When stored in the input fingerprint image memory 19e in the RAM 19, the process proceeds to the third tilt angle calculation process. First, the division axis X1 is set at the center position of the fingerprint image data GS stored in the input fingerprint image memory 19e. Then, the image data is divided into two left and right fingerprint image data GS1, GS2 that are symmetrical about the dividing axis X1 (step E21).
[0076]
Then, a correlation coefficient having the symmetric axis about the division axis X1 is calculated between the left and right fingerprint image data GS1: GS2 divided by the division axis X1, and the angle of the axis X1 and the correlation coefficient value are expressed as the rotation angle / It is stored in the correlation value memory 19l (step E22).
[0077]
Then, while the divided axis X1 is rotated by X2, X3,... By a predetermined angle, the divided axes X2, X3,... Between the left and right fingerprint image data GS1: GS2 divided at the respective rotational positions are centered. A symmetrical correlation coefficient is calculated, and each corresponding split axis angle and correlation coefficient value are stored in the rotation angle / correlation value memory 19l (steps E22 to E24).
[0078]
For example, in the range of ± 90 °, the correlation coefficient between the divided symmetrical fingerprint images GS1: GS2 at the respective rotation angles when the division axis X is rotated by X1, X2, X3,. Calculated and stored in the rotation angle / correlation value memory 19l. When it is determined that the calculation of the correlation value over the entire range of the set rotation angle has been completed (step E23), the calculation is stored in the rotation angle / correlation value memory 19l. Of the correlation values between the divided symmetrical fingerprint image data GS1: GS2 corresponding to the respective rotation angles of the divided axis X, the rotation angle of the divided axis X at which the maximum correlation value is obtained is the input fingerprint image GS. And is stored in the tilt angle memory 19h (step E25).
[0079]
Therefore, according to the fingerprint image collation function of the fingerprint collation device 10 having the above-described configuration, when the fingerprint image data to be registered and the fingerprint image data to be collated are collected, any fingerprint image data GS is matched. However, the inclination angle θ of the fingerprint S in the fingerprint image data GS is obtained, and the fingerprint image data GS is rotationally corrected by the inclination angle θ, so that the registered fingerprint image data without the inclination of the fingerprint S is obtained. Is stored in the registered fingerprint image memory 19f and the verification fingerprint image memory 19g is stored with the verification fingerprint image data when the fingerprint S is not inclined. Then, since the presence or absence of collation match is determined by the comparison correlation between the registered fingerprint image data and the collation fingerprint image data, the collation fingerprint image data is rotated by a predetermined angle during the collation process of the registration fingerprint and the collation fingerprint. However, it is possible to reduce the time from when the verification fingerprint is collected until the determination result is obtained without being burdened with a large processing burden of performing comparison determination with registered image data each time the rotation is performed.
[0080]
Further, according to the fingerprint image collation function by the fingerprint collation apparatus 10 having the above-described configuration, the first fingerprint image captured in the image sensor 17 in a state where no illumination is performed from the light source 18 provided below the fingerprint reading surface H; Since the input fingerprint image data GS is calculated by calculating the difference in shading with the same second fingerprint image captured in the illuminated state, this is collected as registered fingerprint image data or collation fingerprint image data. It is possible to prevent variations in the density of the fingerprint image data every time a fingerprint is collected due to the influence of light wrapping, and to solve problems such as a collation error of the registered person's fingerprint image.
[0081]
Further, according to the fingerprint image collation function by the fingerprint collation device 10 having the above-described configuration, the amount of external light wraparound is determined based on the first fingerprint image captured in the image sensor 17 without illumination from the light source 18; Since the amount of illumination by the light source 18 and the shutter speed (imaging speed) in the image sensor 17 are appropriately variably set and fingerprint image data that is less affected by external light is collected, fingerprint verification with higher reliability can be performed. It can be carried out.
[0082]
In the above-described embodiment, the inclination angle θ of the fingerprint S in the fingerprint image is obtained for any of the registered fingerprint image data and the verification fingerprint image data, and each fingerprint image data is rotationally corrected by the inclination angle θ. Then, it is stored in the registered fingerprint image memory 19f and the collation fingerprint image memory 19g in a state where there is no inclination at the time of fingerprint collection, and the collation judgment process based on the comparison correlation between the images is very difficult without using heavy processing such as image rotation. Although the configuration is simple, the inclination angle θ of the fingerprint S is added as the inclination information of the registered fingerprint image data and the collation fingerprint image data. The rotation correction may be performed so that the inclination angle θ2 of the other fingerprint image matches the inclination angle θ1. Even in this case, since the inclination angle θ in each fingerprint image is clear in advance, it is possible to immediately match the rotation direction of the fingerprint image and perform quick matching processing.
[0083]
It should be noted that the method described in each of the embodiments, that is, the fingerprint data collection process shown in the flowchart of FIG. 4, the light amount control process shown in the flowchart of FIG. 5, and the fingerprint image registration process shown in the flowchart of FIG. 7, the fingerprint image matching process shown in the flowchart of FIG. 7, the first inclination angle calculation process shown in the flowchart of FIG. 8, the second inclination angle calculation process shown in the flowchart of FIG. 10, and the FIG. Each method such as the third inclination angle calculation process shown in the flowchart is a memory card (ROM card, RAM card, etc.), magnetic disk (floppy (R) disk, hard disk, etc.) as programs that can be executed by a computer. , An external recording medium 13 such as an optical disk (CD-ROM, DVD, etc.), semiconductor memory, etc. It can be distributed and paid. The computer of the fingerprint collation device reads the program recorded in the external recording medium 13 into the storage device 12, and the operation is controlled by the read program, whereby the fingerprint image collection described in each of the embodiments is performed. -A collation function can be realized and the same processing can be executed by the method described above.
[0084]
Further, program data for realizing each of the above methods can be transmitted on the communication network N in the form of a program code, and the program is controlled by the communication control unit 14 of the fingerprint collation apparatus connected to the network N. It is also possible to implement the above-described fingerprint image collection / collation function by fetching data.
[0085]
The present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the invention when it is practiced. Further, each of the embodiments includes inventions at various stages, and various inventions can be extracted by appropriately combining a plurality of disclosed constituent elements. For example, even if some constituent elements are deleted from all the constituent elements shown in one embodiment or the constituent elements shown in some embodiments are combined, they are described in the column of the problem to be solved by the invention. In the case where the problems described above can be solved and the effects described in the “Effects of the Invention” can be obtained, a configuration in which these constituent requirements are deleted or combined can be extracted as an invention.
[0086]
【The invention's effect】
As described above, according to the image data collating apparatus according to the present invention, the inclination of the fingerprint in the read fingerprint image data is calculated by the inclination calculating means, and is verified based on the calculated inclination of the fingerprint image data. The inclination of the image data is adjusted. And since the collation fingerprint image data in which the inclination of the fingerprint is matched and the registered fingerprint image data are collated, It is possible to easily perform fingerprint collation with an appropriate fingerprint image without rotational deviation .
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of an electronic circuit of a fingerprint collation apparatus 10 according to an embodiment of an image data collation apparatus of the present invention.
FIG. 2 is a diagram showing a configuration of a fingerprint reading unit in the fingerprint collation apparatus 10;
FIG. 3 is a view showing a data memory secured in a RAM 19 of the fingerprint collation apparatus 10;
FIG. 4 is a flowchart showing fingerprint data collection processing by the fingerprint collation apparatus 10;
FIG. 5 is a flowchart showing a light amount control process accompanying the fingerprint data collection process of the fingerprint collation apparatus 10;
FIG. 6 is a flowchart showing fingerprint image registration processing by the fingerprint collation apparatus 10;
FIG. 7 is a flowchart showing fingerprint image collation processing by the fingerprint collation apparatus 10;
FIG. 8 is a flowchart showing a fingerprint image registration process and a first tilt angle calculation process associated with the fingerprint image collation process by the fingerprint collation apparatus 10;
FIG. 9 is a diagram illustrating a calculation principle of a fingerprint tilt angle θ associated with a first tilt angle calculation process of the fingerprint collation apparatus 10; FIG. 9 (A) is a reading state of a fingerprint S corresponding to a fingerprint reading range H; FIG. 8B is a diagram showing a storage state of the read fingerprint image data GS in the input fingerprint image memory 19e.
FIG. 10 is a flowchart showing a fingerprint image registration process and a second tilt angle calculation process associated with the fingerprint image collation process by the fingerprint collation apparatus 10;
FIG. 11 is a diagram showing a calculation principle of a fingerprint tilt angle θ associated with a second tilt angle calculation process of the fingerprint collation apparatus 10; FIG. 11 (A) shows a fingerprint read corresponding to a fingerprint reading range H; The figure which shows the storage state in the input fingerprint image memory 19e of the image data GS, The figure (B) is the figure which shows the standard fingerprint pattern P from which the correlation coefficient is calculated with the read fingerprint image data GS.
FIG. 12 is a flowchart showing a fingerprint image registration process and a third tilt angle calculation process associated with the fingerprint image collation process by the fingerprint collation apparatus 10;
FIG. 13 is a diagram showing a calculation principle of a fingerprint tilt angle θ associated with a third tilt angle calculation process of the fingerprint collation apparatus 10; FIG. 13 (A) shows a fingerprint read corresponding to a fingerprint reading range H; The figure which shows the storage state in the input fingerprint image memory 19e of the image data GS, The figure (B) (C) is division | segmentation symmetrical image GS1: GS2 when the division | segmentation axis | shaft X is set with the different rotation angle on the reading fingerprint image data GS. The figure which shows the correlation state between.
[Explanation of symbols]
10: Fingerprint verification device
11: Control unit (CPU)
12 ... Storage device
13: External recording medium
14 Communication control unit
15 ... Input device
16 (16a to 16d): Contact sensor
17 Image sensor
18 (18a, 18b) ... light source
19 ... RAM
19a: first fingerprint image memory
19b ... second fingerprint image memory
19c: Light quantity setting memory
19d ... Shutter speed setting memory
19e ... Input fingerprint image memory
19f ... registered fingerprint image memory
19g ... collation fingerprint image memory
19h ... Inclination angle memory
19i ... correlation value memory
19j ... work area
19k ... Ellipse pattern memory
19l ... Rotation angle / correlation value memory
20 ... Display device
21 ... Reading transparent plate
22: Imaging lens
N: Communication network
H ... Fingerprint reading range
GS ... fingerprint image data
S: Fingerprint
J ... central axis of fingerprint
θ… Fingerprint tilt angle
P: Standard ellipse pattern
X: Fingerprint split axis
GS1, GS2 ... divided symmetrical fingerprint image data

Claims (2)

An image data collating device for comparing and collating a fingerprint image registered in advance with a newly read fingerprint image,
The correlation value at each rotation angle is calculated while rotating a predetermined ellipse pattern with respect to the fingerprint image data by a predetermined angle, and the rotation angle of the ellipse pattern at which the maximum correlation value is calculated is determined as the rotation angle of the fingerprint in the fingerprint image data. An inclination calculating means as an inclination;
An inclination correction means for matching the inclination of the image data to be collated based on the inclination of the fingerprint image data calculated by the inclination calculation means;
Image collating means for collating the fingerprint image data and the registered fingerprint image data in which the inclination of the fingerprint is adjusted by the inclination correcting means;
An image data collating apparatus comprising: An image data collation processing program for comparing and collating a fingerprint image registered in advance with a newly read fingerprint image by controlling a computer,
The computer,
The correlation value at each rotation angle is calculated while rotating a predetermined ellipse pattern with respect to the fingerprint image data by a predetermined angle, and the rotation angle of the ellipse pattern at which the maximum correlation value is calculated is the fingerprint value in the fingerprint image data. An inclination calculating means for making an inclination;
An inclination correcting means for matching the inclination of the image data to be collated based on the inclination of the fingerprint image data calculated by the inclination calculating means;
Image collation means for collating the fingerprint image data and the registered fingerprint image data in which the inclination of the fingerprint is adjusted by the inclination correction means,
A computer-readable image data collation processing program designed to function as a computer.

Images