JPH03127192A - Fingerprint collating system - Google Patents

Abstract

PURPOSE:To always execute fingerprint collation at high speed with an optimum scanning pitch by calculating the average pitch of a descending line at every individual and setting the scanning pitch corresponding to the descending line pitch at every individual when a finger print is collated. CONSTITUTION:The subject system is provided with a fingerprint sensor 11 to input the picture of the fingerprint, an A/D converter 12 to convert the output of the fingerprint sensor 11 to digital data, a picture memory 13 to store the output of the A/D converter 12, a feature extraction part 14 to extract the feature information of the fingerprint based on data read out from the picture memory 13 and a dictionary memory 15 to store the feature information prepared by the feature extraction part 14. The average descending line pitch of a fingerprint collating area at every individual is measured in advance and registered together with the fingerprint feature information and when the fingerprint is collated, the scanning pitch is set corresponding to the individual descending line pitch information. Then, the fingerprint information read by the fingerprint sensor with the set scanning pitch is collated with the registered fingerprint information. Thus, the fingerprint collation can be always executed at high speed with the optimum scanning pitch.

Description

[Detailed Description of the Invention] [Summary] This invention relates to a fingerprint matching method that identifies individuals by comparing fingerprint information read by a fingerprint sensor with pre-registered fingerprint information. The purpose is to perform fingerprint matching in the fingerprint matching area.The average ridge pitch of the fingerprint matching area for each individual is measured in advance and registered along with the fingerprint characteristic information. Set the scanning pitch, and the fingerprint information read by the fingerprint sensor at the set scanning pitch,
The configuration is configured to match registered fingerprint information.

[Industrial Application Field] The present invention relates to a fingerprint matching method for identifying an individual by comparing fingerprint information read by a fingerprint sensor with previously registered fingerprint information.

In recent years, as computers have been introduced into popular social systems, system security has become a focus of interest among those concerned. Many questions have been raised about the security of ID cards and passwords, which have been used up until now to verify identity when entering computer rooms or using terminals. Fingerprints, on the other hand, are considered the most effective means of identity verification because they have two major characteristics: they are unique for everyone and they remain unchanged throughout life, and much research and development has been conducted on simple personal identification systems using fingerprints. It is being done.

[Prior Art] In a conventional fingerprint matching system based on pattern matching, a template as shown in Fig. 6 is superimposed on the input screen, and the matching points of the two are scanned while shifting one pixel at a time. The required method has been adopted. FIG. 7 is a diagram showing the scanning state. 1 is an input screen on which a fingerprint pattern read by a fingerprint sensor is displayed. The template 2 is scanned in the direction of the arrow while overlapping the read fingerprint pattern.

However, since this method requires a huge amount of time for verification, it has been proposed to speed up the fingerprint verification process by scanning the template 2 by skipping several pixels at a time. For example, set the matching threshold for pattern matching to 759.
6, the scanning pitch of the template can be roughened to 12.5% of the pitch of the fingerprint ridge pattern. FIG. 8 is an explanatory diagram of ridge pitch, and the distance from one ridge to the next ridge is the ridge pitch. Ridge pitch is 200-6
It falls within the range of 00 μm. The width W of the ridge is 150 ~
It is about 300 μm.

Figure 9 shows the ridge pitch (μm) and the optimal maximum scanning pitch (μm).
It is a figure showing the relationship of (m). The characteristics shown in the figure are for a case where the threshold value of matching degree is 75%. For example, when the ridge pitch is 400 μm, the maximum scanning pitch is about 50 μm, which is about 12.5% from the characteristic diagram.

[Problems to be Solved by the Invention] Generally, the pitch of the ridge pattern of a fingerprint is distributed in the range of 200 μm to 600 μm as described above, but in the conventional example, a constant interlaced scanning pitch was used, When trying to match the scanning pitch to a fingerprint with a thick ridge pitch, the problem was that only the parts with a thin ridge pitch could be aligned properly.

The present invention has been made in view of these problems, and its purpose is to provide a fingerprint matching method that can perform fingerprint matching at high speed at a scanning pitch that is optimal for each individual fingerprint.

[Means for Solving the Problems] FIG. 1 is a flowchart showing the principle of the system of the present invention. In the present invention, the average ridge pitch of the fingerprint matching area is measured in advance for each individual and registered along with the fingerprint characteristic information (step 1).
, Set the scanning pitch according to the individual's ridge pitch information during fingerprint verification (Step 2), and compare the fingerprint information read by the fingerprint sensor with the registered fingerprint information using the set scanning pitch (Step 3). ).

[Operation: The average pitch of the ridges is determined for each individual, and the scanning pitch is set according to the ridge pitch for each individual during fingerprint verification. By doing so, it is possible to perform fingerprint verification at high speed and always at an optimal scanning pitch.

[Example] Hereinafter, an example of the present invention will be described in detail with reference to the drawings.

FIG. 2 is a diagram showing an example of the configuration of a fingerprint verification system implementing the present invention. In the figure, 11 is a fingerprint sensor that manually inputs a fingerprint image, 12 is an A/D converter that converts the output of the fingerprint sensor 11 into digital data, and 13 is an image memory that stores the output of the A/D converter 12. be.

Reference numeral 14 denotes a feature extraction unit (also referred to as a dictionary creation unit) that extracts fingerprint feature information based on the data read from the image memory 13.
It is. Here, the fingerprint characteristic information consists of branches and end points as shown in FIG. Reference numeral 15 denotes a dictionary memory that stores feature information created by the feature extractor 14.

FIG. 4 is a diagram showing an example of the configuration of a dictionary memory, in which position information a,
It is composed of a binarized image and a feature point code C. 1
Reference numeral 6 denotes a ridge pitch measurement unit that measures the ridge pitch based on the fingerprint information read from the image memory 13, and 17 refers to ridge pitch information that stores the average value of the ridge pitches measured by the ridge pitch measurement unit 16. It is a storage part. The dictionary memory 15 and the ridge pitch information storage section 17 are provided within the memory 30.

18 is a pattern matching unit that compares and matches the fingerprint pattern read by the fingerprint sensor 11 and the feature information stored in the dictionary memory 15; one unit receives output data from the ridge pitch information storage unit 17 and performs fingerprint matching; A scanning pitch setting circuit that sets the scanning pitch at the time provides scanning pitch information to the pattern matching section 18. 20 manually inputs personal identification information (ID) into the dictionary memory 15 and the ridge pitch information storage section 17;
This is the D input section. The operation of the circuit configured as described above will be explained as follows.

(During Dictionary Registration) At this time, first, the fingerprint image information of each individual is input, and the average value of the ridge pitch for each individual is calculated and stored in the memory 30 together with the characteristic information. Fingerprint information input from the fingerprint sensor 11 is converted into digital data by the A/D converter 12 and then stored in the image memory 13. The ridge pitch measuring section reads the fingerprint information stored in the image memory 13 and calculates the average value of the ridge pitch.

The ridge pitch of a fingerprint is shorter in the center than in the periphery. Therefore, we decided to measure the ridge pitch at the center. Figure 5 is a diagram showing the measurement position of the ridge pitch.As shown in the diagram, three scanning lines L1 to L3 are run in the center of the fingerprint, and the average value of the ridge pitch is determined for each scanning line. seek. The determined average value of the ridge pitches is stored in the ridge pitch information storage section 17. At the same time, the feature extractor 14 extracts feature information as shown in FIG. 3 from the fingerprint pattern read from the image memory 13 and stores it in the dictionary memory 15.

At this time, personal identification information is also input from the ID input section 20. As a result, each individual's identification information, corresponding fingerprint characteristic information, and ridge pitch are stored in the memory 30.

(During Fingerprint Verification) During actual fingerprint verification, fingerprint information is read from the fingerprint sensor 11 and stored in the image memory 13 in the same manner as during registration. On the other hand, from the ID input section 20, the ID number of the individual to be verified is manually entered into the memory 30. As a result, the characteristic information of the individual is read from the dictionary memory 15 and sent to the pattern matching section 18, and the ridge pitch information storage section 17 outputs the ridge pitch of the individual. The scanning pitch setting circuit 19 receives this ridge pitch information, sets an optimal scanning pitch according to the ridge pitch, and provides the set to the pattern matching section 18 . The scanning pitch at this time is preferably about four times the ridge pitch, for example.

The pattern matching unit 18 interlace scans the image memory at a scanning pitch set in response to this information, and superimposes the template indicating the feature information read from the dictionary memory 15 on the fingerprint pattern read from the image memory 13 to find a match. A matching operation is performed to check the points. According to the invention,
Since the ridge pitch of each individual is measured in advance, matching can be performed while interlaced scanning is performed according to the ridge pitch, resulting in increased speed.

[Effects of the Invention] As described above in detail, according to the present invention, the average pitch of ridges is determined for each individual, and the scanning pitch is set for each individual according to the ridge pitch during fingerprint verification. By doing so, fingerprint verification can be performed at high speed with always an optimal scanning pitch.

[Brief explanation of the drawing]

Fig. 1 is a flowchart showing the principle of the method of the present invention, Fig. 2 is a diagram showing a configuration example of a fingerprint matching system implementing the invention, Fig. 3 is a diagram showing an example of fingerprint characteristic information, and Fig. 4 is a dictionary. Figure 5 shows an example of the memory configuration, Figure 5 shows the ridge pitch measurement position, Figure 6 shows an example of the template, Figure 7 shows the scanning state, Figure 8 shows the ridge pitch. FIG. 9 is a diagram showing the relationship between the ridge pitch and the optimum maximum scanning pitch. In FIG. 2, 11 is a fingerprint sensor, 12 is an A/D converter, 13 is an image memory, 14 is a feature extraction section, 15 is a dictionary memory, 16 is a ridge pitch measurement section, and 17 is a ridge pitch information storage section. , 18 is a pattern matching section, 1 is a scanning pitch setting section, 20 is an ID manual section, and 30 is a memory.

Claims (1)

[Claims] The average ridge pitch of the fingerprint matching area is measured in advance for each individual and registered together with the fingerprint characteristic information (step 1).
, During fingerprint verification, the scanning pitch is set according to the individual's ridge pitch information (Step 2), and the fingerprint information read by the fingerprint sensor is compared with the registered fingerprint information using the set scanning pitch (Step 2). 3) A fingerprint verification method characterized by the following.