JP2886772B2 - Collation device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a collating apparatus for recognizing a pattern using an optical correlation operation.

[0002]

2. Description of the Related Art Since pattern recognition can be used for personal identification, for example, a collation device that uses a fingerprint pattern as personal information and performs collation of individuals based on the fingerprint pattern has been developed. Kaihei 1-2357
6 etc.).

Various methods for performing high-speed pattern correlation by optical calculation have been proposed. For example, a method using a hologram or a joint transform correlation (Joint Transform Correlation: J) method has been proposed.
TC) (“Fourier transform in optical information processing”, Optics 21, 6, 392-399). This joint conversion correlation method uses a Vander-Lugt optical correlator (“pattern recognition using an optical correlation system phase filter”, applied physics).
58,6 (1989)), there is an advantage that accurate positioning of a hologram dry plate on which a pattern is written is not required.

FIG. 9 is a block diagram showing a configuration of a conventional recognition device for collating a fingerprint pattern using a joint conversion correlation method. The recognition apparatus shown in FIG. 1 uses high-resolution spatial light modulators (SLMs) 200 and 201, and converts an input pattern (input image) and a reference pattern (reference image) into a CRT2.
02 are simultaneously drawn to detect the coincidence of the two patterns. The input image and the reference image are written to the SLM 200 and Fourier transformed to obtain a Fourier pattern.
These patterns are read out again by the SLM 201 and subjected to optical Fourier transform. Correlation signal images of these patterns are provided to the photodetector (PD) 203.

When the input image and the reference image match,
In the correlation signal image given to the PD 203, a point of the 0th-order light corresponding to the intensity of the original pattern appears at the center, and two points of ± 1st-order light having high luminance correspond to the positions of the input image and the reference image. The image that appeared in The coincidence between the input pattern and the reference pattern can be detected from the point of ± 1st order light of this image.

[0006]

In the above-described conventional apparatus, the ± first-order light points are signals indicating the correlation intensity between the input pattern and the reference pattern. It changes greatly depending on spatial frequency characteristics. In other words, even if the degree of coincidence of the pattern is 100%, the detected correlation strength greatly differs depending on the type of the pattern.

FIG. 10 shows the result of autocorrelation calculation of a fingerprint pattern of 10 fingers using a conventional recognition device. The autocorrelation calculation is a correlation calculation performed between the same patterns as shown in FIG. From FIG. 10, it can be seen that the autocorrelation value has a variation of about ± 20% from a maximum intensity of about 2500 to a minimum intensity of about 1600. This range becomes even larger as the number of samples increases. This is due to the fact that different fingerprint patterns cause differences in contrast, area, spatial frequency, and the like.

In the conventional collating apparatus, a fixed threshold value is determined in advance, and when a correlation degree equal to or higher than this threshold value is obtained, it is determined that the patterns are the same. However, since the autocorrelation value varies for each pattern as described above, an error may occur in this determination.

SUMMARY OF THE INVENTION It is an object of the present invention to solve such a problem and to provide a matching device capable of detecting a match between an input pattern and a reference pattern with high accuracy and at high speed.

[0010]

A collating apparatus according to the present invention determines whether or not a newly input pattern and one reference pattern read from a plurality of registered reference patterns are the same. Performing an autocorrelation operation for each of the reference patterns, determining a threshold value for each of the reference patterns in accordance with an autocorrelation strength obtained as a result, and setting the reference pattern and the threshold value as a pair, Threshold value setting means for registering a plurality of pairs of a pattern and the threshold value in a memory, and a correlation operation for performing a correlation operation between the input pattern and one of the reference patterns registered in the memory Means, a correlation value between the reference pattern and the input pattern obtained from the correlation calculation means, and a threshold value stored as a pair with the reference pattern. Te, wherein the reference pattern and the input pattern and a determination means for determining whether or not the same. According to the matching device of the present invention, an autocorrelation operation is performed for each reference pattern, and a threshold value for each reference pattern is determined according to the resulting autocorrelation strength,
A plurality of pairs of reference patterns and thresholds are registered in the memory, with one pair of reference patterns and thresholds. Therefore, when it is determined whether or not the newly input pattern and the one reference pattern read out from the plurality of registered reference patterns are the same, the input pattern is previously stored as a pair with the reference pattern. Since the threshold value of the reference pattern can be used, it is possible to quickly and accurately determine whether or not the input pattern and the reference pattern are the same.

Further, the collation device of the present invention is a collation device for determining whether or not a newly input fingerprint pattern and one reference fingerprint pattern read from a plurality of registered reference fingerprint patterns are the same. Performing an autocorrelation operation for each of the reference fingerprint patterns, determining a threshold value for each of the reference fingerprint patterns according to the autocorrelation strength obtained as a result, and taking the reference fingerprint pattern and the threshold value as a pair, Threshold value setting means for registering a plurality of pairs of the reference fingerprint pattern and the threshold value in a memory; and a correlation between the input fingerprint pattern and one of the reference fingerprint patterns registered in the memory. Correlation operation means for performing an operation, a correlation value between the reference fingerprint pattern and the input fingerprint pattern obtained from the correlation operation means, and a pair with the reference fingerprint pattern. By comparing the threshold value stored Te, the input fingerprint pattern and the reference fingerprint pattern is characterized in that it comprises a determining means for determining whether the same. According to the matching device of the present invention, an autocorrelation operation is performed for each reference fingerprint pattern, and a threshold value for each reference fingerprint pattern is determined according to the autocorrelation strength obtained as a result. As a pair, a plurality of pairs of the reference fingerprint pattern and the threshold are registered in the memory. Therefore, when it is determined whether or not the newly input fingerprint pattern and one reference fingerprint pattern read out from the plurality of registered reference fingerprint patterns are the same, a pair with the reference fingerprint pattern is determined in advance. Since the threshold value of the reference fingerprint pattern stored as the reference fingerprint pattern can be used, it is possible to quickly and accurately determine whether or not the input fingerprint pattern and the reference fingerprint pattern are the same.

[0012]

An embodiment of the present invention will be described below with reference to the accompanying drawings. FIG. 1 is a block diagram illustrating the configuration of the matching device according to the present embodiment. This verification device is a device that performs personal verification using a fingerprint pattern of a finger. As shown in the figure, the collating apparatus of the present embodiment detects an input unit 10 for taking in a fingerprint input pattern, a memory 20 in which a fingerprint reference pattern and threshold data are registered, and detects a correlation value between the input pattern and the reference pattern. And a correlation operation unit 30 that performs the operation. Further, a threshold setting unit 40 for setting a threshold level based on a reference pattern registered in the memory 20.
A determination unit 50 that determines whether the input pattern and the reference pattern are the same pattern from the correlation value data output from the correlation calculation unit 30, a correlation calculation unit 30, a threshold setting unit 40, and a determination unit 50. And a control unit 60 for controlling.

The input unit 10 includes a prism 11, a light emitting diode (LED) 12 for irradiating the prism 11 with light,
A CCD camera 13 for capturing an input pattern of a fingerprint;
Lens 14 for imaging an input pattern on a CD camera 13
And

The correlation operation section 30 includes a CRT 301 for displaying an input pattern and a reference pattern side by side, a spatial light modulator (SLM) 302 of an optical address type for converting the input pattern and the reference pattern into a coherent optical image, and a coherent optical image. And a Fourier transform lens 303 that obtains a joint Fourier transform optical image by Fourier transforming Also, an optical address type SLM 304 for converting a joint Fourier transform light image into a coherent light image, a Fourier transform lens 305 for Fourier transforming the coherent light image to obtain a correlation signal image, and a photo detector (PD) for detecting the correlation signal image 306. Further, between the SLM 302 and the Fourier transform lens 303, the read laser beam 310
SLM3
04 and the Fourier transform lens 305 between the SLM 304
Mirror 31 that gives read laser beam 310 to
3,314.

Next, the operation of this embodiment will be described.
The input of the fingerprint pattern uses the total reflection input method.
Light from the ED 12 is applied to the prism 11, and a fingerprint pattern is obtained from the difference in reflection conditions when the verification target person presses his / her finger against the prism 11.

First, when a person to be verified presses a verification start key (not shown), a verification start signal is sent to the control unit 60.
And the collation processing is started. By operating a keyboard (not shown) to input, for example, a personal identification number, a reference pattern registered in advance in the memory 20 is selected, and this reference pattern is sent to the control unit 60.
The selection of the reference pattern may be performed by inserting an ID card into the card reader, instead of using the keyboard. Next, when the verification target person presses his / her finger against the prism 11, the fingerprint pattern obtained on the surface of the prism 11 forms an image on the CCD camera 13. The CCD camera 13 converts the fingerprint pattern into a video signal and sends it to the control unit 60.

The control unit 60 gives the input pattern and the reference pattern sent to the CRT 301 and displays them side by side on the screen of the CRT 301 as shown in FIG. 2A (the left half of FIG. Input pattern, right half is reference pattern). The input pattern and the reference pattern displayed on the CRT 301 correspond to the S pattern
It is written to the LM 302.

The two patterns written on the SLM 302 are read by the read laser beam 310 and are Fourier transformed by the Fourier transform lens 303. And
A joint Fourier transform image as shown in FIG.
304 is formed on the a-Si layer. In addition, SLM3
04 is read out by the laser light 310 and subjected to Fourier transform by the Fourier transform lens 305. In this Fourier transform, the central zero-order light point as shown in FIG.
A correlation signal image composed of the ± 1st light points on both sides of the next light is obtained. The light intensity of the −1st order light point (or + 1st order light point) in the correlation signal image is detected by the PD 306. The light intensity at the -1 order light spot corresponds to the correlation between the two patterns. This PD
The detection signal from 306 is transmitted to the determination unit 50 via the control unit 60.
And threshold processing is performed using the threshold data read from the memory 20 to determine whether the input pattern and the reference pattern are the same pattern. The determination result thus obtained is used, for example, for opening and closing a door and unlocking a computer.

As described above, if the reference pattern of the fingerprint is different, a difference occurs in contrast, area, spatial frequency, etc., and the degree of autocorrelation differs for each reference pattern. For this reason, in this embodiment, threshold data used for comparing an input pattern with a reference pattern is set for each reference pattern and stored in the memory 20. The setting of the threshold data is performed by the threshold setting unit 40 when the reference pattern of the fingerprint is registered in advance by the person to be verified.

Conventionally, as disclosed in Japanese Patent Application Laid-Open No. 63-149776, attempts have been made to determine the coincidence rate for each individual. However, in the case of the present embodiment using the optical correlator, the detected correlation intensity and the coincidence rate do not correspond directly (correspondence intensity is greatly different even when the pattern is exactly the same, that is, the coincidence rate is 100%). . Therefore, a unique threshold setting method for the optical correlator is required.

In this embodiment, a method of setting a threshold value based on an autocorrelation value and a method of setting a threshold value based on an autocorrelation value and a cross-correlation value are used. The processing procedure of these setting methods will be described with reference to the flowcharts of FIGS. FIG. 3 is a flowchart illustrating a procedure for setting threshold data based on the autocorrelation value. As shown in the figure, first, when a person to be verified presses a finger on the prism 11 of the input unit 10 to register a fingerprint reference pattern, the fingerprint reference pattern is read by the CCD camera 13 (step 70). The read reference pattern is stored in the memory 2 under the control of the control unit 60.
0 is registered (step 71).

The read reference pattern is provided to a threshold value setting unit, and the reference pattern is copied (step 72). The copied reference pattern and the original reference pattern are provided to the correlation operation unit 30, and the correlation operation between these patterns is performed (step 73). An intensity signal of the autocorrelation value is obtained by the correlation operation, and the intensity signal is supplied to the threshold value setting unit 40. Threshold setting section 4
At 0, a threshold value is determined for each reference pattern based on the intensity signal (step 74). The data of the threshold value determined in this way is controlled according to the control of the control unit 60.
It is registered in the memory 20 (step 75).

As a method of determining the threshold value, when the collation apparatus of the present embodiment is used for a high security application such as access to a bank computer or entry / exit of a safe, the threshold value is set to, for example, an autocorrelation value. 80% of the If the threshold value is set at such a high level, the probability of erroneously determining a person other than the registrant as a registrant is almost zero. On the other hand, when the collation apparatus of the present embodiment is used for a low security application such as entering and leaving a general room, the threshold value is set to, for example, an autocorrelation value of 40.
%. If a threshold value is set at such a low level, although a slight erroneous determination is made, there is almost no determination that the registrant is not a registrant. As described above, the threshold value setting unit 40 sets the threshold value according to the security level.

FIG. 4 is a flowchart showing a processing procedure for setting threshold data based on the auto-correlation value and the cross-correlation value. In this example, the use of the cross-correlation value in addition to the auto-correlation value is based on the fact that the cross-correlation value of a fingerprint pattern differs from pattern to pattern as well as the auto-correlation value. That is, in a fingerprint pattern including many features in common with other fingerprint patterns, a high-intensity cross-correlation is output. Therefore, the threshold value is determined in consideration of the cross-correlation value in addition to the autocorrelation value, thereby obtaining a more accurate threshold value.

In this process, first, in order for the person to be collated to register the reference pattern of the fingerprint, the prism 1 of the input unit 10 is used.
When a finger is pressed on the reference number 1, a reference pattern of a fingerprint is read by the CCD camera 13 (step 80). The read reference pattern is registered in the memory 20 under the control of the control unit 60 (step 81).

The read reference pattern is supplied to a threshold value setting unit, and the reference pattern is copied (step 82). The copied reference pattern and the original reference pattern are provided to the correlation operation unit 30, and a correlation operation is performed between these patterns to detect an autocorrelation value (step 83).

Further, another registered reference pattern is read from the memory 20 (step 84), and the reference pattern read in step 80 and the other reference pattern are given to the correlation calculator 30. The correlation operation section 30 performs a correlation operation between these patterns, and detects a cross-correlation value (step 85). The autocorrelation value detected in step 83 and the cross-correlation value detected in step 85 are both provided as intensity signals to threshold setting section 40. The threshold setting unit 40 determines a threshold for each reference pattern based on these intensity signals (step 86). The data of the threshold value thus determined is registered in the memory 20 under the control of the control unit 60 (step 87).

Next, how to set a threshold value in the threshold value setting section 40 will be described with reference to experimental results shown in FIGS. FIGS. 5 to 7 are graphs showing the results of experiments in which the auto-correlation value and the cross-correlation value of a fingerprint pattern are detected by a correlation operation. In each graph, experiments are performed using reference patterns of different fingerprints (fingerprints A to C). The horizontal axis of these graphs indicates the reference pattern number, and the vertical axis indicates the correlation strength when the autocorrelation value is 1.0. First, in the graph of FIG. 5, the reference pattern of pattern number 0 is the same pattern as the reference pattern of the fingerprint A to be measured, and the correlation value with this pattern is the autocorrelation value. The fingerprint reference patterns of pattern numbers 1 to 29 are different from the fingerprint A pattern, and the correlation value with these patterns is the cross-correlation value. According to this graph, pattern numbers 6 and 18
The cross-correlation value between the No. reference pattern and the reference pattern to be measured becomes maximum at 17% of the auto-correlation value. FIG.
In the graph, the reference pattern of pattern number 1 is the same pattern as the reference pattern of the fingerprint B to be measured, and the correlation value with this pattern is the autocorrelation value. The correlation value with the reference pattern other than the first reference pattern is the cross-correlation value. From this graph, the cross-correlation value between the reference patterns of pattern numbers 4, 6, and 29 and the reference pattern to be measured becomes maximum at 19% of the auto-correlation value. Further, in the graph of FIG. 7, the reference pattern of pattern number 2 is the same as the reference pattern of the fingerprint C to be measured, and the correlation value with this pattern is the autocorrelation value. The correlation value with the reference pattern other than the second reference pattern is the cross-correlation value. According to this graph, the cross-correlation value between the reference pattern of pattern number 5 and the reference pattern to be measured becomes maximum at 22% of the auto-correlation value.

The threshold value needs to be determined from the maximum value of the cross-correlation, the auto-correlation value, the size of the detector, noise, the number of registered fingerprints, the security factor, and the like. The maximum value of 17% (for fingerprint A) is set as the lower limit of the threshold, and the autocorrelation value is set as the upper limit of the threshold. Which level the threshold value is set in the range from the upper limit to the lower limit is determined according to the purpose. For example, when the collation device of the present embodiment is used for a low security application such as entering and leaving a general room, for example, the threshold value is set to a level of 10% of the range between the upper limit and the lower limit. In the case of fingerprint A, the threshold value is (100% −17%) × 0.1 + 17% = 25.3%. On the other hand, in the case of high security such as bank computer access, for example, 70% of the range between the upper limit and the lower limit is set as the threshold level. Thus, the threshold value may be determined according to the purpose.

As described above, by knowing the upper limit level of the expected correlation value from the autocorrelation value and obtaining the lower limit level of the threshold value from the cross-correlation value, the range (safety factor, security degree) can be obtained from this range. By setting the threshold value accordingly, a stable recognition system can be constructed.

In the present embodiment, the threshold value is set by the threshold value setting unit at the timing when the reference target registers the fingerprint reference pattern in advance, but is limited to this timing. Instead, for example, it may be performed at the timing when the reference target person inputs the input pattern of the fingerprint to the input unit 10.

Further, in the present embodiment, the personal identification is performed using the fingerprint, but the present invention is not limited to the fingerprint, and a palm pattern or a sole pattern may be used.

Further, in this embodiment, the SLMs 302, 3
Although an optical address type SLM is used as 04, an electrical address type SLM may be used.

Further, instead of the CRT 301, another display device such as a liquid crystal display or a plasma display may be used.

Further, the fingerprint input by the input unit 10 may be holographic as shown in FIG. This sensor is
A hologram 101 is formed on a part of a transparent glass flat plate (light guide plate) 100, and a laser light is emitted from a laser light source 102 on the opposite side of the other end.
At 3, a fingerprint pattern is detected.

[0036]

According to the collating apparatus of the present invention, it is determined whether or not the newly input fingerprint pattern and the one reference fingerprint pattern read out from the plurality of registered reference fingerprint patterns are the same. When determining whether the input fingerprint pattern and the reference fingerprint pattern are the same, the threshold value of the reference fingerprint pattern stored in advance as a pair with the reference fingerprint pattern can be used. And accurately.

Further, according to the collation device of the present invention, it is determined whether or not the newly input fingerprint pattern and one reference fingerprint pattern read out from the plurality of registered reference fingerprint patterns are the same. When determining whether the input fingerprint pattern and the reference fingerprint pattern are the same, the threshold value of the reference fingerprint pattern stored in advance as a pair with the reference fingerprint pattern can be used. And accurately.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a matching device according to an embodiment.

FIG. 2 is a diagram showing an input image, a Fourier image, and a correlation signal image.

FIG. 3 is a flowchart showing a processing procedure of a threshold value setting method.

FIG. 4 is a flowchart showing a processing procedure of a threshold value setting method.

FIG. 5 is a diagram showing an experimental result of detecting an autocorrelation value and a cross-correlation value by a correlation operation.

FIG. 6 is a diagram showing an experimental result of detecting an auto-correlation value and a cross-correlation value by a correlation operation.

FIG. 7 is a diagram showing an experimental result of detecting an auto-correlation value and a cross-correlation value by a correlation operation.

FIG. 8 is a diagram showing an input unit using holography.

FIG. 9 is a block diagram illustrating a configuration of a conventional recognition device.

FIG. 10 is a diagram showing a result of performing autocorrelation calculation of a fingerprint pattern using a conventional recognition device.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 10 ... Input part, 11 ... Prism, 12 ... Light emitting diode, 13 ... CCD camera, 14 ... Lens, 20 ... Memory, 30 ... Correlation calculation part, 40 ... Threshold value setting part, 50 ...
Judgment unit, 60 ... Control unit, 100 ... Glass flat plate, 101 ...
Hologram, 102: laser light source, 103: image sensor,
301 ... CRT, 302, 304 ... spatial light modulator, 30
3,305: Fourier transform lens, 306: photodetector, 310: laser beam, 311 to 314: half mirror.

──────────────────────────────────────────────────続 き Continuing from the front page (72) Inventor Yuji Kobayashi 1126, Nomachi, Ichinomachi, Hamamatsu City, Shizuoka Prefecture Inside (72) Inventor Tsutomu Hara 1126, Nomachi, Ichinomachi, Hamamatsu City, Shizuoka Prefecture Hamamatsu Photonics Co., Ltd. (56) References JP-A-59-144982 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) G06T 7/00

Claims (5)

(57) [Claims] 1. A collating apparatus for judging whether a newly inputted input pattern and one reference pattern read from a plurality of registered reference patterns are the same, wherein an autocorrelation operation is performed for each of the reference patterns. Performing a threshold value for each of the reference patterns in accordance with the autocorrelation strength obtained as a result, and setting a plurality of pairs of the reference pattern and the threshold value as a pair of the reference pattern and the threshold value. Threshold setting means for registering in the memory; correlation calculating means for performing a correlation operation between the input pattern and one of the reference patterns registered in the memory; A correlation value between the reference pattern and the input pattern is compared with the threshold value stored as a pair with the reference pattern, and the input pattern and the reference pattern are compared. Determining means for determining whether the turns are the same or not. 2. The threshold value setting means further performs a cross-correlation operation between the reference pattern and another registered reference pattern, and performs a calculation for each of the reference patterns according to the auto-correlation strength and the cross-correlation strength. The collation device according to claim 1, wherein a threshold value is determined. 3. A collation device for judging whether or not a newly input fingerprint pattern and one reference fingerprint pattern read from a plurality of registered reference fingerprint patterns are the same. An autocorrelation operation is performed, and a threshold for each of the reference fingerprint patterns is determined according to the autocorrelation strength obtained as a result. The reference fingerprint pattern and the threshold are paired, and the reference fingerprint pattern and the threshold are determined. Threshold setting means for registering a plurality of threshold value pairs in a memory; correlation calculating means for performing a correlation operation between the input fingerprint pattern and one of the reference fingerprint patterns registered in the memory; A correlation value between the reference fingerprint pattern and the input fingerprint pattern obtained from the correlation calculating means, and a correlation value stored as a pair with the reference fingerprint pattern. Determining means for comparing a threshold value with the input fingerprint pattern to determine whether or not the input fingerprint pattern is the same as the reference fingerprint pattern. 4. The threshold setting means further performs a cross-correlation operation between the reference fingerprint pattern and another registered reference fingerprint pattern, and calculates the reference fingerprint pattern according to the auto-correlation strength and the cross-correlation strength. 4. The collating apparatus according to claim 3, wherein a threshold value is determined for each of the groups. 5. The correlation calculation performed by the threshold value setting means and the correlation calculation means using a joint conversion type optical correlator. The collation device according to claim 1.