JPH05205035A - Fingerprint collating device - Google Patents

Abstract

PURPOSE:To provide a fingerprint collating device which can surely eliminate pseudo feature points. CONSTITUTION:A fingerprint collating device consists of a fingerprint sensor 1 which reads the fingerprint images, a fingerprint image storage part 7 which stores temporarily the fingerprint image data read by the sensor 1, a gradation image correcting part 10 which calculates the contour lines of the gradation image of finger prints from the fingerprint image data read by the sensor 1, extracting the pseudo feature points of the is islands, the bridges, the sweat glands, etc., out of an image consisting of those calculated contour lines to reduce the density of an area corresponding to a closed curve enclosed by the contour lines with the feature points showing the islands or the bridges and to increase the density of an area corresponding to a closed curve enclosed by the contour lines with the feature points showing the sweat glands respectively, and then excludes these pseudo feature points out of the gradation image, and a main body device 20 which carries out the binarization, the registration in a dictionary, the fingerprint collation, etc., to the image corrected by the part 10.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a fingerprint collation device.
With the spread of electronic computers in society in recent years, public attention has been focused on how to secure security. I have been used so far as a means of verifying identity when entering a computer room or using a terminal.
Many questions have been raised regarding the security of D cards and personal identification numbers. On the other hand, fingerprints are "universal",
Since it has the two major characteristics of "lifetime immutability", it is considered to be the most effective means for personal identification, and much research and development has been conducted on a simple personal identification system using fingerprints.

[0002]

2. Description of the Related Art FIG. 10 is a conceptual diagram showing the structure of a conventional fingerprint collation device. First, the operation at the time of registration will be described. The fingerprint sensor 1 is pressed against a finger to detect a fingerprint pattern, and is converted into digital data by an A / D converter (not shown) in the fingerprint sensor 1. The converted digital data (fingerprint data) is processed by the subsequent binarization circuit 2.
It is converted into binary data of “0” and “1” and stored in the binarization memory 3.

The fingerprint data stored in the binarized memory 3 is sequentially read out and then the characteristic information is extracted. Here, the characteristic information means, for example, a branch point as shown in FIG. 11A, an end point as shown in FIG. The fingerprint can be specified by the number of such branch points and end points at which positions. The extracted feature information is stored in the fingerprint dictionary storage unit 5
Stored in. The above operation is repeated for a plurality of individuals, and the characteristic information of the individuals is stored in the fingerprint dictionary storage unit 5.

When the registration of the characteristic information in the fingerprint dictionary storage unit 5 is completed in this way, the collation operation of the individual fingerprint is started this time. In the case of collation, a finger (a finger used for registration in advance; for example, an index finger) is placed on the fingerprint sensor 1, and then an ID number of the user is input with a ten-key pad (not shown). As a result,
The collation unit 6 can narrow the search range by determining the search range of the fingerprint dictionary storage unit 6 based on the ID number and reading it out at the time of matching.

The fingerprint pattern is detected and converted into digital data by the A / D converter in the fingerprint sensor 1 in the same manner as the registration. The converted digital data is converted into binary data of “0” and “1” by the subsequent binarization circuit 2 and stored in the binarization memory 3.

The collation unit 6 reads out the fingerprint image for collation stored in the binarization memory 3 and the characteristic information for each individual stored in the fingerprint dictionary storage unit 5, and collates both (pattern matching). I do. If the number of matching feature patterns is greater than or equal to a predetermined number, it is determined that the fingerprints match.

Here, when the characteristic information extracting section 4 extracts the characteristic information, it is necessary to eliminate pseudo characteristic points such as fingerprint bridges and sweat glands. This is because if such a pseudo feature point is used for fingerprint collation, an erroneous operation occurs in the identification of the person.

FIG. 12 is a diagram showing a conventional example of a sweat gland removing method. First, as shown in (a), a thin line 31 and a branch point 32.
Although there are a and 32b, an image, 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 (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 predetermined number of pixels, the thin line is formed. To be regarded as. This is because in the case of sweat glands, the fine line is always looped, and the size of the closed curve is within the predetermined range as a hole for sweat, so there is no error even if it is uniformly detected. Is.

Within this predetermined number of pixels is used for the determination as the size of the closed curve of the sweat gland. In the case of (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, and is shown in (c). As described above, these branch points 32a and 32b are removed from the feature points.

[0011]

The above-described method for removing pseudo-feature points is based on the mutual relationship between the thin line and the feature points. However, the pseudo-feature points are often left and the true feature points are removed. There was a problem of removing it.

The present invention has been made in view of the above problems, and an object of the present invention is to provide a fingerprint collation device capable of reliably removing pseudo feature points.

[0013]

FIG. 1 is a block diagram showing the principle of the present invention. The same parts as those in FIG. 10 are shown with the same reference numerals omitted. In the figure, 1 is a fingerprint sensor for reading a fingerprint image, 7 is a fingerprint image storage unit for temporarily storing the fingerprint image data read by the fingerprint sensor 1, and 10 is a fingerprint from the fingerprint image stored in the fingerprint image storage unit 7. The contour lines of the grayscale image of are calculated, and pseudo feature points such as islands, bridges or sweat glands are extracted from the image composed of these contour lines, and when these feature points are islands or bridges, a closed curve surrounded by these contour lines is created. A density image correction unit that reduces the density of the corresponding points, while increasing the density of the points corresponding to the closed curve surrounded by these contour lines when these feature points are sweat glands, and eliminates these pseudo feature points from the gray image, Is a main body device that performs processing such as binarization, dictionary registration, and fingerprint collation on the image corrected by the grayscale image correction unit 10.

[0014]

The grayscale image correction unit 10 calculates contour lines of the grayscale image of the fingerprint from the fingerprint image stored in the fingerprint image storage unit 7, and from the image composed of these contour lines, pseudo features such as islands, bridges or sweat glands. Points are extracted, and if these feature points are islands or bridges, the density of the part corresponding to the closed curve surrounded by these contour lines is reduced, while if these feature points are sweat glands, the closed curve surrounded by these contour lines is reduced. The density of the corresponding part is increased to eliminate these pseudo feature points from the grayscale image. Then, the main body device 20 performs processing such as binarization, dictionary registration, and fingerprint collation on the image from which the pseudo feature points have been removed. With such a configuration, it is possible to provide a fingerprint collation device that can reliably remove pseudo feature points.

[0015]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 2 is a configuration block diagram showing an embodiment of the present invention. The same parts as those in FIG. 1 are designated by the same reference numerals. In the figure, 1 is a fingerprint sensor, 7 is a fingerprint image storage unit for storing a fingerprint image read by the fingerprint sensor 1, and 10 is a fingerprint image storage unit.
Calculates contour lines of a grayscale image of a fingerprint from the fingerprint image stored in the fingerprint image storage unit 7, and extracts pseudo feature points such as islands, bridges or sweat glands from the image constituted by these contour lines,
When these feature points are islands or bridges, the concentration of the closed curve enclosed by these contour lines is reduced, while when these feature points are sweat glands, the concentration of the closed curve enclosed by these contour lines is reduced. Is a grayscale image correction unit that eliminates these pseudo feature points from the grayscale image, and is a feature of the present invention.

Reference numeral 21 is a binarization circuit for binarizing the fingerprint image from which the pseudo feature points are removed, and 22 is the binarization circuit 2
A binarized image storage unit for storing data binarized by 1;
Reference numeral 23 is a thinning circuit for performing thinning processing on the image data stored in the binarized image storage unit 22, and 24 is a thinning image storage for storing the fingerprint image data thinned by the thinning circuit 23. 25 is a feature extraction circuit for extracting feature points from the image data stored in the thinned image storage unit 24, and 26 is a feature point extracted by the feature extraction circuit 25 and image data in the vicinity thereof. Window selection circuit that cuts out as a window,
A dictionary output circuit 27 outputs the window image selected by the window selection circuit 26 as a feature point dictionary. Here, the components 21 to 27 described above constitute the main body device 20 of FIG. In addition, the arrow shown by the solid line in the figure indicates the flow of data,
The arrow shown by a broken line shows a signal flow. The operation of the apparatus configured as described above will be described below.

Fingerprint image data read from the fingerprint sensor 1 is stored in the fingerprint image storage unit 7. The grayscale image correction unit 10 performs the work of removing the pseudo feature points from the image data (grayscale image) stored in the fingerprint image storage unit 7. Hereinafter, the operation of the grayscale image correction unit 10 will be described in detail.

The gradation image correction unit 10 includes a fingerprint image storage unit 7
From the grayscale image stored in, the location where the density is constant is detected. FIG. 3 is an explanatory diagram of contour line detection. (A)
Indicates the scanning direction. Scan horizontally or vertically.
Then, with respect to a certain threshold value TH, a portion showing the same density is detected. (B) shows the detection principle of a pixel having a threshold value TH. The vertical axis represents the density and the horizontal axis represents the scanning direction.
Since the grayscale image has discrete values, the pixels that exactly match the threshold TH are not always on the threshold TH.

If the pixel is above the threshold TH, and if the pixel is not above the threshold TH, the pixel closest to the threshold TH is selected. That is, the density A (I, Y)
When <TH, A (I + 1, Y)> TH is satisfied,
Either (I, Y) or (I + 1, Y) is adopted. By setting this constant value in several ways, it is possible to obtain contour lines composed of a plurality of layers.

FIG. 4 is a diagram showing an example of the obtained contour lines.
(A) shows the relationship between the island cross section and the contour line, and (b) shows the relationship between the sweat gland cross section and the contour line. A plan view and a cross-sectional view taken along the line AA are shown respectively.
The island shown in (a) has a high center, and the sweat gland shown in (b) has a low center.

FIG. 5 is an explanatory diagram of a closed curve detecting method.
The obtained contour lines are traced so as to check whether or not to return to the starting point within the threshold number of pixels. When such a closed curve is obtained, for example, when it is clockwise, it is −9 with respect to the direction vector.
Draw a perpendicular in the direction of 0 ° to see if there is another closed curve inside.

In a bridge, adjacent ridges adhere to each other, and in many cases, an island in which a part of a valley line is raised is often seen to adhere to the surrounding ridges. FIG. 6 is an explanatory diagram of bridge generation. (A) is a binarized image, (b) is its contour line, (c) is its bird's-eye view. For this reason, the bridge has a higher density than the valley line, and when binarized it becomes the same "1" as the ridge. However, the concentration may be lower than that at the center of the ridge or the island, and in such a case, a closed curve can be obtained by appropriately selecting the height of the contour line. The closed curve formed by the bridges and islands has a higher concentration on the inner closed curve (see Fig. 4). Also,
Also in the case of sweat glands, a closed curve is obtained by contour lines, but in this case, the density is lower in the closed curve inside (see FIG. 4).

In this way, the grayscale image correction unit 10 detects a portion corresponding to a pseudo feature point such as a bridge, an island, or a sweat gland in the grayscale image, and stores the closed curve. And
Since it is possible to distinguish between a bridge and an island and a sweat gland under the above conditions, the concentration in the closed curve is lowered in the former case, and is increased in the latter case.

FIG. 7 is a diagram showing a state of density correction.
(A) shows density correction for islands, and (b) shows density correction for sweat glands. As shown in the figure, the density correction may be performed on a region outside by one or two pixels from the closed contour curve. This is because, for example, in the case of a bridge, the contour line of the adhesion portion of the island and the ridge does not always become a closed curve. The value set by the correction is set to a value lower than the outer closed curve L in the former case and higher than the outer closed curve L in the latter case. By correcting the grayscale image in this manner, the pseudo feature points can be removed without using the positional relationship between the thin line and the feature points.

In this way, the image for which the removal of pseudo feature points is completed is stored again in the fingerprint image storage unit 7. The binarization circuit 21 binarizes the fingerprint image stored in the fingerprint image storage unit 7. The binarized image data is stored in the binarized image storage unit 22. Thinning circuit 23
Performs a thinning process on the image data stored in the binarized image storage unit 22, and stores the thinned image storage unit 24 in the thinned image storage unit 24.

The feature extraction circuit 25 extracts feature points from the fingerprint image data for which the thinning process has been completed. Since the grayscale image correction unit 10 has completed the removal of the grayscale image indicating the pseudo feature points before the binarization, the feature points extracted here do not include the pseudo feature points. The window selection circuit 26 cuts out a window consisting of the feature points extracted by the feature extraction circuit 25 and the vicinity thereof, and outputs the window from the dictionary output circuit 27. This output window image becomes a fingerprint dictionary.

FIG. 8 is a block diagram showing an example of a concrete configuration of the grayscale image correction section 10. In the figure, a fingerprint image storage unit 7 stores a grayscale image of a fingerprint as digital data. The contour line detection circuit 11 reads out data from the fingerprint image storage unit 1, sets several thresholds TH, adopts pixels having a density of TH, or when the density of consecutive pixels is before and after TH, By adopting one of the pixels, the contour line storage unit 12 is used as a pixel of the contour line of the density TH.
To memorize. The contour line storage unit 12 stores the number of contour line images corresponding to the number of threshold values TH.

The contour line tracking circuit 13 includes a contour line storage unit 12.
The image data stored in is read and the contour lines are tracked. Since contour lines are generally derived irrespective of the edge of the screen, untraced contour lines are detected and traced by the number of pixels within the threshold value, similarly to the scanning shown in FIG. The inner closed curve tracking circuit 14 detects a closed curve as shown in FIG. 5 with respect to the tracked contour lines. That is, when returning to the starting point with the number of pixels within the threshold value, the closed curve indicating the island, the edge, the sweat gland, etc., and the densities of these pixels are stored in the closed curve and density storage unit 15. The method shown in FIG. 5 is used to examine the density of a closed curve inside a certain closed curve.

The correction area correction density setting circuit 16 corrects the density of the closed curve of the island, the bridge, and the sweat gland by the method as shown in FIG. However, the area to be corrected is set to be wider by 1 or 2 pixels than the outermost closed curve, and the correction value is a ridge,
Use the average density of the valley line.

FIG. 9 is a flow chart showing the operation of the present invention. The figure shows the operation at the time of registration. First, a fingerprint image is input from the fingerprint sensor 1 (S1), and it is checked whether all the threshold values have been completed (S2). If not, the threshold value is set to TH (S3), and scanning is performed in the X direction. Then, it is checked whether or not the scanning in the X direction is completed (S4), and if it is completed, the process returns to step S2 to check whether or not all the threshold values are completed.

If not completed, the contour detection circuit 11 checks whether A (X-1, Y) <TH and A (X, Y) ≧ TH (1) (S5). When the expression (1) is satisfied, (X, Y) at that time is set as a new contour line (S7), and the process returns to step S4. When the expression (1) is not satisfied, the contour line detection circuit 11 checks whether A (X, Y) ≧ TH and A (X + 1, Y) <TH (2) is satisfied (S6). When the expression (2) is satisfied, (X, Y) at that time is set as a new contour line (S7), and the process returns to step S4. Even if the formula (2) is not satisfied, the process returns to step S4.

Next, in step S2, if the processing has been completed for all threshold values, it is checked whether the processing has been completed for all contour lines (S8). If the processing is not completed, the contour line tracing circuit 13 traces the contour lines (S9). Then, it is checked whether or not the number of tracking pixels ≧ threshold value (3) (S10), and if so, the process returns to step S8. If the equation (3) does not hold, it is checked whether the starting point has been reached (S
11). If not reached, the process returns to step S9 to trace the contour lines. When it reaches, the inner closed curve detection circuit 14 stores the coordinates and the density of the closed curve in the closed curve and density storage unit 15 (S12).

In step S8, when the calculation of the coordinates and the density of the closed curve for all contour lines is completed,
The correction area correction density setting circuit 16 performs density correction (S
13), and the result is stored again in the fingerprint image storage unit 7.

When the fingerprint image from which the pseudo feature points have been removed is stored in the fingerprint image storage unit 7, the binarization circuit 21 (see FIG. 2) performs binarization (S14), and then the thinning circuit 23. Thinning is performed (S15). A feature point is extracted by the feature extraction circuit 25 from the fingerprint image that has been thinned (S1).
6), output as a dictionary from the dictionary output circuit 27 (S1)
7).

In the above description, the case of scanning in the X direction as the scanning direction is taken as an example, but the present invention is not limited to this. The scanning may be performed in the Y direction as shown in FIG. The density correction value can also be the density of contour lines outside the closed curve.

[0036]

As described above in detail, according to the present invention, it is possible to provide the fingerprint collation device capable of surely removing the pseudo feature points.

[Brief description of drawings]

FIG. 1 is a principle block diagram of the present invention.

FIG. 2 is a block diagram showing an example of the overall configuration of the present invention.

FIG. 3 is an explanatory diagram of contour line detection.

FIG. 4 is a diagram showing an example of contour lines.

FIG. 5 is an explanatory diagram of a method for detecting a closed curve.

FIG. 6 is an explanatory diagram of bridge generation.

FIG. 7 is a diagram showing a state of density correction.

FIG. 8 is a block diagram illustrating a specific configuration example of a grayscale image correction unit.

FIG. 9 is a flowchart showing the operation of the present invention.

FIG. 10 is a conceptual diagram of a configuration of a conventional fingerprint matching device.

FIG. 11 is a diagram showing an example of fingerprint characteristic information.

FIG. 12 is a diagram showing a conventional example of a sweat gland removal method.

[Explanation of symbols]

 1 Fingerprint Sensor 7 Fingerprint Image Storage 10 Grayscale Image Correction 20 Main Unit

Claims (5)

[Claims] 1. A fingerprint sensor (1) for reading a fingerprint image
A fingerprint image storage unit (7) for temporarily storing fingerprint image data read from the fingerprint sensor (1), and contour lines of a grayscale image of the fingerprint from the fingerprint image stored in the fingerprint image storage unit (7). Calculate and extract pseudo feature points such as islands, bridges or sweat glands from the image composed of these contour lines, and if these feature points are islands or bridges, calculate the density of the location corresponding to the closed curve enclosed by these contour lines. On the other hand, when these feature points are sweat glands, on the other hand, the density of the portion corresponding to the closed curve surrounded by these contour lines is increased, and the grayscale image correction unit (10) for eliminating these pseudo feature points from the grayscale image, and the grayscale image. 2 for the image corrected by the correction unit (10)
Main unit that performs processing such as digitization, dictionary registration, fingerprint matching (2
0) and a fingerprint matching device. 2. The grayscale image correction unit (10), when calculating a contour line, determines a certain threshold value and repeats the process of connecting a series of grayscale images substantially equivalent to the threshold value with a line for the required number of threshold values. The fingerprint collation apparatus according to claim 1, wherein the contour line is created by doing so. 3. When the density image correction unit (10) determines an island, a bridge, a sweat gland, etc. in a contour image, if the density of the closed curve surrounded by the contour is higher on the outer side than on the inner side, it is determined as a sweat gland. The fingerprint collation device according to claim 1, wherein the determination is made, and conversely, when the density is lower on the outside than on the inside, it is determined to be an island or a bridge. 4. The fingerprint according to claim 1, wherein the gray-scale image correction unit (10) sets the area to be density-corrected outside the outermost closed curve by 1 or 2 pixels. Matching device. 5. The fingerprint collation according to claim 1, wherein the grayscale image correction unit (10) determines the value after density correction by the value of a contour line outside the closed curve to be corrected. apparatus.