JPH05242227A - Fingerprint collating device - Google Patents

Abstract

PURPOSE:To provide the fingerprint collating device which shortens the time required for collation even at the time of using a directional filter to improve the picture. CONSTITUTION:This device consists of a first binarizing circuit 10 which reads in and binarizes a fingerprint picture, a register part 20 which divides the output of the binarizing circuit 10 into plural blocks and extracts the direction of rising lines in each block to register a fingerprint dictionary, a second binarizing circuit 12 which binarizes the output of a fingerprint sensor 1 after using a directional filter corresponding to the direction of rising lines in each block detected by the register part 20 to eliminate bridges and cracks of rising lines of fingerprint, a feature extracting circuit 13 which extracts feature point information of the output of the binarizing circuit 12 to give it to the register part 20, a fingerprint dictionary register part 30 where feature point information, window picture data of the periphery of feature points, and rising line direction data of a pertinent window picture are received and registered from the register part 20, and a collating part 40 which collates the picture for collation from the binarizing circuit 12 with the fingerprint dictionary from the fingerprint dictionary storage part 30.

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. 8 is a conceptual diagram showing the construction 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. 9 (3a), an end point as shown in FIG. 9 (b), or the like. 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. When the number of matching feature patterns is equal to or more than a predetermined number, it is determined that the fingerprints match (moving window method).

Next, the moving window method will be described in detail. From the fingerprint ridge image as shown in FIG.
As shown in 1, a plurality of windows including the feature point and its vicinity are cut out. Of the cut out windows, the window in the middle is the alignment window. The window shown in FIG. 11 is registered as a dictionary.
That is, the fingerprint dictionary is divided into one alignment window and a plurality of matching windows.

Next, at the time of matching, the matching fingerprint image is aligned with the positioning window, and pattern matching with the matching window is performed as shown in FIG. First, the registration window is scanned over the entire fingerprint image, pattern matching is performed, and the positional deviation amount of the input image with respect to the registered image is grasped. After that, the collation window is displaced by the displacement amount of the alignment window and collation is performed. At this time, in order to cope with the distortion caused by the softness of human skin, the matching window is two-dimensionally scanned by a slight amount to perform pattern matching.

FIG. 13 is a flowchart showing the operation of the conventional device. First, pattern matching of the alignment window image is performed (S1). In this pattern matching, as shown in FIG. 12, the position matching window is superimposed on the read fingerprint image so that both images match. Then, it is checked whether or not it is a candidate point depending on whether or not the positioning window can be matched (S2).

If it is not a candidate point, the positioning window is moved (S3) and it is checked whether it is within the action range (S4). If it is within the action range, the process returns to step S1. If it is outside the range of action, the fingerprint collation is rejected (S5).

If a candidate point is found in step S2, pattern matching is performed on the other matching window (matching feature point) images by the same amount as the movement amount of the positioning window (S6). Then, it is checked whether the pattern matching has passed (S7). If you pass,
Add 1 to the number of pattern matchings that have passed until then,
The number of passes is compared with the reference threshold (S8). When the number of passes ≧ threshold value, the person is confirmed (S9).

If it fails in step S7, it is checked whether all windows are finished (S10),
When all windows are completed, it means that the number of passes does not reach the reference threshold value, and therefore it is refused to be judged as the person (S11). If the pattern matching for all windows has not been completed, the process moves to another peripheral image (S12) and pattern matching is performed. It should be noted that step S10 is also executed when the number of passes is not greater than or equal to the threshold value in step S8.

[0013]

The fingerprint collation is performed by observing the coincidence of the distribution and shape of the characteristic points of the fingerprint as described above. However, since the fingerprint input unit generally uses the optical contact between the finger and the glass, if the finger is sweating, a bridge due to sweat is formed between the fingerprint ridges, and the fingerprint ridge pattern looks like a branch. The fingerprint image is detected as if the dots were present. On the contrary, when the finger is dry, the contact between the finger and the glass becomes poor and the fingerprint ridge is cut in the fingerprint image.

Therefore, a large number of fingerprint end points appear in the fingerprint image, and when the feature points are extracted, the cut points of the fingerprint ridges may be extracted as the feature points. In order to eliminate such inconvenience, there is a method of extracting the direction of the fingerprint ridges or valleys of the detected fingerprint image and removing the bridges or cracks of the fingerprint ridges with a directional filter along the ridge direction. , The direction of fingerprint ridges is extracted during registration and verification,
If you use a directional filter to improve the image,
There is a drawback that matching takes time.

The present invention has been made in view of the above problems, and provides a fingerprint collation device capable of reducing the time required for collation even when an image is improved by using a directional filter. It is an object.

[0016]

FIG. 1 is a block diagram showing the principle of the present invention. In the figure, 1 is a fingerprint sensor for reading a fingerprint image, 10 is a first binarization circuit for binarizing the output of the fingerprint sensor 1, and 20 is an output of the first binarization circuit 10. A registration unit that divides each block into a plurality of blocks, extracts the direction of the ridge for each block, and registers the fingerprint dictionary. Reference numeral 11 receives the output of the fingerprint sensor 1 and performs multi-valued image processing using a directional filter corresponding to the ridge direction of each block detected by the registration unit 20 to detect fingerprint ridge bridges and cracks. A filter unit 12 for removing is a second binarization circuit for receiving the output of the filter unit 11 and binarizing it.

A reference numeral 13 receives the output of the second binarization circuit 12, extracts characteristic information, and outputs the result to the registration unit 20.
A feature extraction circuit 30 for giving the feature point information output from the feature extraction circuit 13 and its vicinity as a window,
A fingerprint dictionary registration unit that receives and registers the data output as the feature point window image and the ridge direction data of the window image from the registration unit 20, and 40 is a collation image output from the second binarization circuit 12. And a fingerprint dictionary output from the fingerprint dictionary storage unit 30.

[0018]

In the registration unit 20, the direction data for each block of the binarized fingerprint image is obtained and given to the filter unit 11. The filter unit 11 performs image processing in that direction based on the direction data provided from the registration unit 20, and removes images other than that direction, such as bridges and cracks. The binarization circuit 12 binarizes the image from which the bridges and cracks are removed, and the feature extraction circuit 13 extracts feature points. The extracted feature points and their peripheral information are stored as windows in the fingerprint dictionary storage unit 30 from the registration unit 20. Furthermore, the direction data used to determine the direction of the ridge is also block-compatible with the fingerprint dictionary storage unit 30.
The registration unit 20 stores the information.

With this configuration, even if the collation fingerprint image is input, the collation unit 40 compares the direction data in the target block of the collation fingerprint image with the direction data of the alignment window image. By doing so, it is only necessary to compare the direction data, so that the alignment can be performed quickly.
Further, if the alignment based on the direction data does not match, it is possible to determine that the matching does not match without performing the pattern matching, so that the fingerprint matching becomes faster.

[0020]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT An embodiment of the present invention will now be described in detail with reference to the drawings. FIG. 2 is a block diagram showing the structure of an embodiment of the present invention. The same parts as those in FIG. 1 are designated by the same reference numerals. In the figure, the fingerprint sensor 1 is composed of a fingerprint image input section 1a and an A / D converter 1b. In the registration unit 20, 21 is a direction data extraction unit that divides the binarized data into blocks and extracts direction data for each block.
Reference numeral 2 denotes a window image creation unit that receives an output of the feature extraction circuit 13 and cuts out an image including a feature point and its vicinity as a window.
Is a window image direction data creation unit that receives the direction data from the direction data extraction unit 21 and the feature point window image from the window image creation unit 22 and creates the window image direction data.

The fingerprint dictionary storage unit 30 is composed of window image direction dictionary data 31, alignment window image dictionary data 32, and matching peripheral window dictionary data 33. Collation unit 4
In 0, 41 is a direction data collating unit that collates the direction data stored in the fingerprint dictionary storage unit 30 with the direction data of the collation image read from the binarization circuit 12, and 42 is stored in the fingerprint dictionary storage unit 30. An alignment unit for aligning the registered alignment window image with the matching image read from the binarization circuit 12, 43 is a peripheral matching window image stored in the fingerprint dictionary storage unit 33 and 2 It is a peripheral window matching unit that performs matching with the matching image read from the binarization circuit 12. The operation of the apparatus thus configured will be described below with reference to the flowchart.

FIG. 3 is a flow chart showing the operation of the first embodiment of the present invention, showing the operation at the time of collation. In order to perform the collating operation, the dictionary data needs to be stored in the fingerprint dictionary storage unit 30. Therefore, the operation at the time of registration will be described. The fingerprint image input from the fingerprint image input unit 1a is converted into multivalued image data by the subsequent A / D converter 1b. This multi-valued data is binarized by the binarization circuit 10 and then enters the registration unit 20.

In the registration unit 20, the direction data extraction unit 21 receives the output of the binarization circuit 10, divides it into a plurality of blocks, and extracts the direction of the ridge for each block. FIG. 4 is a diagram showing an example of ridge direction data. The binarized fingerprint image shown in (a) is divided into blocks, and the direction of the ridge is extracted for each divided block. The extracted result is as shown in (b).

The direction data for each block extracted by the direction data extraction unit 21 is sent to the filter unit 11. Multi-valued image data that has not been binarized is input to the filter unit 11. The filter unit 11 performs processing for removing noise such as bridges and cracks in the block based on the direction data for each block provided from the direction data extraction unit 21. FIG. 5 is an explanatory view of correcting bridges and cracks by the directional filter. Even if the image contains cracks or bridges as shown in (a), the bridges and cracks are removed as shown in (b) by filtering in the ridge direction with the directional filter unit 11. Becomes

The multi-valued fingerprint image corrected by the filter unit 11 is binarized by the binarization circuit 12. Then, after the fingerprint image data binarized by the binarization circuit 12 is thinned, the feature extraction circuit 13 performs feature extraction. The extracted feature points are provided to the window image creation unit 22. The window image creation unit 22 cuts out a window including the input feature points and images in the vicinity thereof. Direction data and window image creation unit 22 extracted by the direction data extraction unit 21
The window image created in 3 is given to the window image direction data creation unit 23. Then, the pixel window image direction data creation unit 23
Sends the input window image and the direction data for the window to the fingerprint dictionary storage unit 30. In the fingerprint dictionary storage unit 30,
Input the fingerprint dictionary data to the window image direction dictionary data 3
1, the window image dictionary data 32 for alignment and the peripheral window dictionary data 33 for matching are stored therein.

After the fingerprint dictionary is registered in this way, the fingerprint collation operation is started. When the fingerprint image for collation is input from the fingerprint sensor 1 (S1), multivalued data is output from the fingerprint sensor 1. This multivalued data is binarized by the binarization circuit 10 and then the direction data extraction unit 21.
to go into. The direction data extraction unit 21 divides the input binary image into a plurality of blocks and extracts the direction of the ridge for each block (S2).

The filter unit 11 includes a direction data extraction unit 21.
Based on the direction data output from, the filter process in that direction is performed to remove bridges and cracks (S3). For the multi-valued image from which bridges and cracks have been removed, the binarization circuit 10
The value is converted (S4). The direction data matching unit 41 selects a position matching area by comparing the direction data of the position matching window image read from the window image direction dictionary data 31 with the direction data of the matching fingerprint image output from the binarization circuit 12. Is performed (S5).

When the matching by the direction data matching unit 41 coincides, the position matching window image read from the position matching window image dictionary data 32 in the region selected next by the position matching unit 42 and the binarization circuit 12 are read. Pattern matching is performed with the matching window image read from (S6). If this matching is matched, the peripheral window matching unit 43 then compares the matching peripheral window image read from the matching peripheral window dictionary data 33,
Pattern matching with the binary image read from the binary circuit 12 is performed (S7). Then, as a result of the pattern matching, if the matching error is equal to or less than the threshold value, the person is confirmed. As described above, according to the present invention, even when the image is improved by using the directional filter, the alignment is first performed using the directional data, so that the time required for the matching can be shortened. FIG. 6 is a flow chart showing the operation of the second embodiment of the present invention. Figure 3 shows the sequence until the fingerprint dictionary is registered.
The operation at the time of matching will be described. When the verification fingerprint image is input from the fingerprint sensor 1 (S
1), multi-valued data is output from the fingerprint sensor 1. This multi-valued data is binarized by the binarization circuit 10 and then enters the direction data extraction unit 21. The direction data extraction unit 21 divides the input binary image into a plurality of blocks,
The direction of the ridge for each block is extracted (S2).

The filter unit 11 includes a direction data extraction unit 21.
Based on the direction data output from, the filter process in that direction is performed to remove bridges and cracks (S3). For the multi-valued image from which bridges and cracks have been removed, the binarization circuit 10
The value is converted (S4). The direction data matching unit 41 compares the direction data of the alignment window image read from the window image direction dictionary data 31 with the direction data of the matching fingerprint image output from the binarization circuit 12 (S5).

By this comparison operation, the direction data collating unit 4
1 checks if the orientations of both images match. If they do not match, the process returns to step S5, and the comparison operation of both direction data is restarted. If both direction data match, then the direction data of the matching window image and the ridge direction data in each block of the matching window fingerprint image are compared as shown in FIG. 7 (S7). .. As shown in (a), it is assumed that there are six ridge direction data in the matching window image. The six ridge direction data are compared with the direction data of the collation fingerprint image as shown in (b).

When a predetermined number of matches are obtained as a result of the comparison, in the collating unit 40, the aligning unit 42 uses the aligning window (thick □ in FIG. 7) to make the aligning window. Pattern matching is performed (S9). That is, the alignment window image dictionary data 32 is read out and the binarization circuit 1 is read.
The pattern matching with the matching window image read out from 2 is performed. If the predetermined number of matches cannot be obtained in step S8, the process returns to step S5 and the direction data of the alignment window is compared again.

In step S9, after the pattern matching of the positioning window image is performed, the deviation from the origin position at this time is obtained. Then, the peripheral window matching unit 43 reads the matching peripheral window dictionary data 33 and performs pattern matching with the matching fingerprint image (S10). If the number of matching pixels of the window image by pattern matching is equal to or more than the threshold value, the matching unit 40 confirms that the person is the person.

Next, a third embodiment of the matching method of the present invention will be described. It is assumed that the fingerprint dictionary storage unit 30 stores the respective dictionary data in the window image direction dictionary data 31, the alignment window image dictionary data 32, and the matching peripheral window dictionary data 33. First, a filter corresponding to the ridge direction extracted in each block is used to correct the multivalued image in the filter unit 11 to remove bridges and cracks. Next, the alignment unit 40 causes the alignment unit 42 to perform pattern matching on the alignment window image of the dictionary.

Next, at the position where the alignment window coincides, the direction data collating unit 41 causes the ridge direction data of the collation fingerprint image and the ridge line of the collation window image at the place where the alignment window coincides. Confirm that the direction data match.
When the ridge directions of the number of windows equal to or more than the threshold value match, the peripheral window matching unit 43 performs pattern matching of the matching window. The collation unit 40 confirms the identity of the person when the number of coincidences of the window images is equal to or larger than the threshold value.

[0035]

As described above in detail, it is possible to provide a fingerprint collation device capable of shortening the time required for collation even when an image is improved by using a directional filter.

[Brief description of drawings]

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

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

FIG. 3 is a flowchart showing the operation of the first exemplary embodiment of the present invention.

FIG. 4 is a diagram showing an example of ridge direction data.

FIG. 5 is an explanatory diagram for correcting bridges and cracks by using a directional filter.

FIG. 6 is a flowchart showing the operation of the second exemplary embodiment of the present invention.

FIG. 7 is a diagram showing comparison of ridge direction data of matching window images.

FIG. 8 is a conceptual diagram showing the configuration of a conventional fingerprint matching device.

FIG. 9 is a diagram illustrating an example of fingerprint characteristic information.

FIG. 10 is a diagram showing an example of a fingerprint ridge image.

FIG. 11 is a diagram showing an example of a fingerprint dictionary image.

FIG. 12 is a diagram showing matching between a collation image and a dictionary.

FIG. 13 is a flowchart showing the operation of the conventional device.

[Explanation of symbols]

 1 Fingerprint Sensor 10 Binarization Circuit 11 Filter Section 12 Binarization Circuit 13 Feature Extraction Circuit 20 Registration Section 30 Fingerprint Dictionary Storage Section 40 Matching Section

Claims (4)

[Claims] 1. A fingerprint sensor (1) for reading a fingerprint image
And a first binarization circuit (12) for binarizing the output of the fingerprint sensor (1), and receiving the output of the first binarization circuit (12) and dividing into a plurality of blocks. , The ridge direction of each block is extracted, and the registration unit (20) that registers the fingerprint dictionary and the output of the fingerprint sensor (1) receive the ridges of each block detected by the registration unit (20). The multi-valued image is processed using the directional filter corresponding to the line direction, and the bridge of fingerprint ridges,
A filter unit (11) for removing cracks, and a second unit for receiving the output of the filter unit (11) and performing binarization.
And a feature extraction circuit (13) which receives the output of the second binarization circuit (12) and extracts feature information, and gives the result to the registration unit (20). And the feature point information output from the feature extraction circuit (13) and its vicinity are cut out as a window, and the data output as a feature point window image and the ridge direction data of the window image are received from the registration unit (20). Fingerprint dictionary registration unit (30)
And a collation unit (40) for collating the collation image output from the second binarization circuit (12) with the fingerprint dictionary output from the fingerprint dictionary storage unit (30). apparatus. 2. The collation unit (40), when performing fingerprint collation, when collating the alignment window image with the collation fingerprint image, first, the direction data in the block of interest in the collation image, The fingerprint collation according to claim 1, wherein the pattern matching of the alignment window image is started only when the direction data of the alignment window image is compared with each other and a match equal to or more than a threshold value is found. apparatus. 3. The collation unit (40), when collating a registration window image with a collation fingerprint image when performing fingerprint collation, first aligns with the direction data in the target block in the collation image. When the comparison is made with the direction data of the window images for a match or more, and the direction data of the matching window images are compared, and a direction match is found for a predetermined number or more of the matching window images. 2. The fingerprint collation apparatus according to claim 1, wherein pattern matching between the position alignment window image and the collation window image is started only in step 1. 4. The matching unit (40) performs pattern matching of a positioning window image with a matching fingerprint image, and if the matching error is less than or equal to a predetermined value, then the matching window image orientation data is acquired. 2. The fingerprint collation apparatus according to claim 1, wherein the pattern collation of the collation window image is started when the direction data of a certain number or more of the window images are coincident with each other.