JP2551631B2 - Personal verification device - Google Patents

Description

Detailed Description [Table of Contents] Outline Industrial field of application Conventional technology (Fig. 7) Problem to be solved by the invention Means for solving the problem Action Example One embodiment of the present invention (first example) ~ Fig. 6) Effect of the invention [Outline] The present invention relates to an individual collation device, which aims to provide an individual collation device capable of accurately identifying branch points and sweat glands to improve fingerprint registration and collation accuracy. After the fingerprint image is input as an image and the fingerprint image is thinned, the feature points including the branch points are extracted while removing the sweat glands on the ridges, and the fingerprint of the individual is registered and collated based on the feature points. In the personal verification device, a sweat gland removing means for removing the sweat gland is provided, and the sweat gland removing means draws a vertical bisector between any two branch points of the thinned image, and the vertical bisector. Is less than the specified number of pixels. If it intersects with the line, it is regarded as a sweat gland, and these two bifurcation points are removed from the feature points.

[Industrial applications]

The present invention relates to a personal verification device, and more particularly, to a personal verification device for identifying an individual using a fingerprint.

With the spread of computers throughout society in recent years, public attention has been focused on how to secure safety. Many questions have been raised regarding the security of ID cards and personal identification numbers that have been used up to now as means of personal identification when entering the computer room or using terminals. On the other hand, fingerprints are "universal"
-Since it has the two major characteristics of "lifetime immutability", it is considered to be the most powerful means of personal identification, and much research and development has been conducted on a simple personal identification system using fingerprints.

[Conventional technology]

In the development of a personal collation device that uses fingerprints, a "real-time fingerprint sensor" that can clearly input the fingerprint image of the person in real time and the registered feature information and the input fingerprint image are practically sufficient. Both of the "algorithms" that perform high-speed matching while maintaining high reliability are key.

With the "real-time fingerprint sensor", it is necessary to input a clear fingerprint image without distortion without being affected by the residual fingerprints of the previous user. The present inventor has developed a "holographic fingerprint sensor" having a new structure in which a reflective light guide plate is combined.

On the other hand, in the development of the "algorithm", it is necessary to flexibly deal with the difference in how force is applied during input and the distortion of the fingerprint image caused by moving with the finger on the glass. The present inventor has newly developed a "moving wind algorithm" that has both the redundancy of the so-called "partial pattern matching method" and the structural analysis of the "feature extraction method".

As a concrete prior art by the above development, the present inventor prototyped a simple personal collation device consisting of a 16-bit personal computer with a frame memory and a matching circuit and an independently developed "holographic fingerprint sensor" to verify the identity. We have confirmed high-speed matching of about 0.7 seconds and about 1.5 seconds for excluding others.

[Problems to be Solved by the Invention]

However, in such a conventional individual collation device, when there is another branch point near a certain branch point in the thinned image for which feature extraction has been completed, this is directly removed as a sweat gland. Therefore, even if there are two true bifurcation points, this removes them, and there is a problem in that the accuracy of fingerprint registration and verification decreases.

For example, in FIG. 7, the vicinity of the feature points of the fingerprint image is cut out as a dotted line frame 1. In this figure, 2 indicated by a solid line is a ridge of a fingerprint (so-called convex portion corresponding to a mountain),
B represents the branch point at the intersection of the descending line 2. Now, if there is another branch point 4 near the branch point 3, these branch points 3 and 4 are simply removed as sweat glands, which is one of the noises. However, since the branch points 3 and 4 are actually the true branch points of the ridge, it is desired to accurately identify them. On the other hand, in the case of a true sweat gland, it is necessary to clearly identify this as a branch point.

Therefore, an object of the present invention is to provide a personal collation device that can accurately identify a branch point and a sweat gland to improve the accuracy of fingerprint registration and collation.

[Means for solving the problem]

In order to achieve the above-mentioned object, the individual collation device according to the present invention inputs a fingerprint of a specific individual, thins the fingerprint image, extracts feature points including branch points while removing sweat glands from ridges, In an individual collation device for registering and collating the fingerprint of the individual based on feature points, a sweat gland removing unit for removing the sweat gland is provided, and the sweat gland removing unit is provided between any two branch points of the thinned image. A vertical bisector is drawn to, and if the vertical bisector intersects two fine lines within a predetermined number of pixels, it is regarded as a sweat gland, and these two branch points are removed from the feature points. .

[Action]

In the present invention, a vertical bisector is drawn between any two branch points in the thinned image, and if the vertical bisector intersects the two thin lines within a predetermined number of pixels, it is considered as a sweat gland. , These two branch points are removed from the feature points. In this case, the sweat glands always form a closed curve, and the size thereof is substantially uniform, so that the sweat glands can be reliably detected.

 Therefore, the branch point and the sweat gland can be accurately identified.

〔Example〕

 Hereinafter, the present invention will be described with reference to the drawings.

1 to 6 are views showing an embodiment of an individual collation device according to the present invention.

First, the configuration will be described. FIG. 1 is a block diagram showing the function of the personal collation device. In this figure, the personal collation device is constructed by using, for example, a 16-bit personal computer, and is roughly divided into an image input system 11 and an image storage circuit 1.
2. It has a function represented by a binarization / thinning circuit 13, a feature extraction circuit 14, a feature point storage device 15, a branch point search circuit 16, a vertical bisector calculation circuit 17, and a sweat gland determination circuit 18.

The image input system 11 comprises, for example, a fingerprint sensor of a type in which a hologram is used as a light source and a hologram and a flat light guide plate are combined, and the fingerprint is received by an image pickup element and converted into image input. The image storage circuit 12 receives the image input from the image input system 11 as A /
The image is stored by performing D conversion or the like, and the stored data is taken out to the outside as needed, is sent to the feature point storage device 15, and the data is exchanged with the binarization thinning circuit 13. Transfers are made.

The binarization and thinning circuit 13 binarizes the image data and thins it into ridges and valleys. The feature extraction circuit 14 extracts feature points from the thinned image, and the feature point storage device 15 stores the feature points. The branch point search circuit 16 searches for a branch point from the thinned image, the vertical bisector calculation circuit 17 draws a vertical bisector between any two branch points, and the sweat gland determination circuit 18 further determines the vertical line. If the bisector intersects with two fine lines within a certain number of pixels, it is regarded as a sweat gland, a process of removing these two branch points from the feature points is performed, and the result is output to the feature point storage device 15.

The data of the above image processing is sent from the feature point storage device 15 or the image storage circuit 12 to an external CRT display or the like and monitored. Further, in the image processing, for example, the processing necessary for creating a window, searching for a feature point, a branch point, etc. is performed using a ten-key pad, a mouse, or the like.

Then, the fingerprint of the individual is registered and the registered data is compared with the image input data to collate the fingerprint. The vertical bisector 17 and the sweat gland determining circuit 18 correspond to sweat gland removing means 19.

 Next, the operation will be described.

FIG. 2 is a flow chart showing a program for fingerprint registration in the personal verification process, and P _{1 to} P _{10 in the} figure show steps of the program.

First, enter the fingerprint image of the specific individual to be registered in the P _1. In this embodiment, an image size of 256 × 256 pixels is used. Then, binarizing the input image P _2. This is, for example, in order to deal with the grayscale distribution of the image due to the imbalance of finger pressing at the time of input, the image is divided into small regions of 16 × 16 pixels, and the average density of the grayscale image is obtained for each small region. It is binarized by using it as a threshold. This is called local binarization. The image is decomposed into two digital values of "1" and "0" by this binarization process, but the image is output as an actual image on the CRT.

Next, a thinning process is performed in P ₃ , and the fingerprint pattern is divided into bright ridges (convex peaks) and valleys (concave valleys). At P ₄ , feature extraction is performed.
For example, a feature extraction mask 21 consisting of 9 pixels as shown in FIG. 3 is provided, and a central point of interest 22 (corresponding to the central pixel) is white 23 in the remaining 8 pixels in the vicinity (part of the circle).
When there is only one or two in succession, it is regarded as an "end point". In addition, when three or more whites 23 independently exist in the neighboring 8 pixels, this central pixel is set as a “branch point”.
And Such feature extraction processing is performed for all branch points, but in P ₅ , it is determined whether or not all branch points have been extracted. If NO, the process proceeds to P ₆ , and if YES, the result is obtained in P _7. Is output and the routine ends.

In P ₆ , it is determined whether or not there is a branch point in the neighborhood, and when it is not a neighborhood, it returns to P _5, and when it is in the neighborhood, vertical bisector processing is performed in P ₈ , and in P ₉ this vertical bisector is processed. It is determined whether there is an intersection that intersects with the line. When there is an intersection was removed regarded as sweat glands branch points corresponding with P _10, when there is no intersection returns to P _5. When the above process is completed for all the branch points in P ₅ , the process proceeds to P ₇ and the result is output.

Step P ₈ to P ₁₀ is a part for realizing the functions of the sweat glands removing means 19 will be specifically described with reference to Figure 4. First, a thin line 31 as shown in FIG. 4 (a) and branch points 32a, 3
For an image having 2b, that is, a thinned image for which feature extraction has been completed, another branch point 32b in the vicinity of one branch point 32a is detected by spiral scanning as indicated by an arrow 33. Then
As shown in FIG. 4 (b), a vertical bisector 33 is drawn between the two branch points 32a and 32b, and if it intersects with two thin lines within a certain number of pixels, it is regarded as a sweat gland. . This is because in the case of sweat glands, the fine line is always looped (closed curve), and the size of the closed curve is within a predetermined range as a hole for sweat, so even if it is detected uniformly, it is incorrect. Because there is no.

The number of pixels within the predetermined number is used to determine the size of the closed curve of the sweat glands. In the case of FIG. 4 (b), the predetermined number of pixels corresponds to the distance NA, and the vertical bisector 33 intersects the fine line 31 in both directions within the range of NA, so it is regarded as a sweat gland,
As shown in FIG. 4 (c), these branch points 32a and 32b are removed from the feature points.

FIG. 5 shows an example of the sweat gland removal process. In FIG. 5, first, one branch point + is detected, and these branch points +
In the vicinity of, the spiral scanning is sequentially performed as indicated by the broken line, and another branch point + is detected. Next, the center C between the two branch points + is calculated, and a vertical bisector ◯ passing through the center C is drawn. When the number of pixels of the vertical bisector ○ is counted from the center C in each of two directions and a point on the thin line is reached within a certain number of pixels NA (intersection X corresponds to this), a sweat gland is present here. Considered to be. Then, the two branch points + are removed from the feature points, and then the screen scanning is resumed, and the above series of operations are repeated for other branch points.

FIG. 6 is a diagram for explaining an effect when the present invention is applied to an image similar to the conventional example. In this figure, branch points 3 and 4 on the ridge 2 are detected, a vertical bisector 41 is drawn between the branch points 3 and 4, and the number of intersections is counted. Here, the only intersection is the point 42, and no closed curve is formed on the vertical bisector 41. Therefore, it is considered not to be a sweat gland, and the branch points 3 and 4 are not removed. Therefore, the branch point and the sweat gland can be accurately identified, and the accuracy of fingerprint registration can be improved.

It should be noted that the program of the present embodiment is a process at the time of fingerprint registration, but when a fingerprint is collated to judge an individual, naturally the branch point and the sweat gland can be accurately identified to improve the accuracy of fingerprint collation. Of course you can.

[Effect] According to the present invention, the branch point and the sweat gland can be accurately identified with respect to the thin line of the fingerprint, and the accuracy of fingerprint registration and verification can be improved.

[Brief description of drawings]

1 to 6 are diagrams showing an embodiment of an individual collation apparatus according to the present invention, FIG. 1 is a functional block diagram thereof, FIG. 2 is a flowchart showing a program of the fingerprint registration processing, and FIG. The figure which shows the feature extraction mask, FIG. 4 is the figure explaining the sweat gland removal, FIG. 5 is the figure which shows the process of that sweat gland removal, FIG. 6 is the figure which explains the process about the branch point, FIG. The figure is a diagram for explaining branch point processing in a conventional individual collation device. 2 ... Ridge line, 3, 4, 32a, 32b ... Branch point, 11 ... Image input system, 12 ... Image storage circuit, 13 ... Binarization thinning circuit, 14 ... Feature extraction circuit, 15 ...... Feature point storage device, 16 …… Branch point search circuit, 17 …… Vertical bisector calculation circuit, 18 …… Sweat gland judgment circuit, 19 …… Sweat gland removal means, 33, 41 …… Vertical bisector .

Claims (1)

(57) [Claims] 1. A fingerprint of a specific individual is input as an image, the fingerprint image is thinned, feature points including branch points are extracted while removing sweat glands on the ridge, and the fingerprint of the individual is extracted based on the feature points. In a personal verification device for registering and verifying fingerprints, a sweat gland removing means for removing the sweat glands is provided, and the sweat gland removing means draws a vertical bisector between any two branch points on the thinned image. A personal collation device characterized in that if the vertical bisector intersects two fine lines within a predetermined number of pixels, it is regarded as a sweat gland, and these two branch points are removed from the feature points. .