JP3452732B2 - Fingerprint 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 fingerprint collation device for collating fingerprints based on spatial frequency characteristics.

[0002]

2. Description of the Related Art In recent years, in fields requiring personal recognition such as entry / exit control for computer rooms and important machine rooms, access control for computer terminals and bank financial terminals,
A voice collating device and a fingerprint collating device are being adopted in place of the personal identification number and the ID card used so far.

The present applicant has previously filed Japanese Patent Application No. 7-108526.
"Pattern matching device" was proposed as the issue. In this pattern matching apparatus, the matching fingerprint image data is created by performing a two-dimensional discrete Fourier transform on the image data (two-dimensional pattern data) of the matching fingerprint. Then, the matching Fourier image data is combined with the registered Fourier image data of the registered fingerprint created by performing the same processing, and amplitude suppression processing (log processing) is performed on the combined Fourier image data. A two-dimensional discrete Fourier transform is applied.
Then, upper n pixels having a higher correlation component intensity than a predetermined correlation component area appearing in the synthesized Fourier image data subjected to the two-dimensional discrete Fourier transform are extracted, and the correlation component intensity of the extracted n pixels is extracted. The correlation value is the average of
Compare with threshold. If the correlation value is higher than the threshold value, it is determined that the registered fingerprint and the verification fingerprint match.

[0004]

However, in this pattern matching apparatus, information indicating the generality of fingerprints such as patterns (information commonly seen in fingerprints) and information indicating individuality such as feature points are different. We handle this in the same way,
It was not possible to say that the recognition rate was sufficient.

The present invention has been made to solve such a problem, and an object thereof is to suppress common information of fingerprints and enhance or extract characteristic information to improve collation accuracy . It is to provide a fingerprint collation device capable of doing the above.

[0006]

In order to achieve such an object, the first invention (the invention according to claim 1) is a registration finger.
Create a registration Fourier image data by performing two-dimensional discrete Fourier transform on the image data of the fingerprint, by performing two-dimensional discrete Fourier transform for the collation fingerprint image data to create a collation Fourier image <br/> data, registration For the Fourier image data and the matching Fourier image data, the frequency zero is individually set as the starting point.
Pixels in the range of surrounding x1 to x2 (0 <x1 <x2)
The same weighting process that emphasizes the data of
By suppressing the common information of the fingerprint and the collation fingerprint, the registered Fourier image data and the collation Fourier image data that have been subjected to this weighting processing are combined, and the combined Fourier image data obtained by this are combined. By performing either one of the two- dimensional discrete Fourier transform and the two- dimensional discrete inverse Fourier transform on each individual data that constitutes the image data of the correlation component area appearing in the Fourier-transformed composite Fourier image data. The registered fingerprint and the collated fingerprint are collated based on the strength of the correlation component.

According to the present invention, the registered Fourier image data and the matching Fourier image data are individually divided into their peripheries.
Surroundings x1 to x2 (0 <x1 <x2) starting from wave number zero
Weighting process to emphasize the data of pixels within the range
The common information of the registered fingerprint and the collated fingerprint is suppressed ,
The registered Fourier image data subjected to this weighting process and the matching Fourier image data are combined, and the two- dimensional discrete Fourier transform or the two- dimensional discrete inverse Fourier transform is applied to the combined Fourier image data obtained by this. . Then, based on the intensity of the correlation component for each individual data that constitutes the image data of the correlation component area that appears in the composite Fourier image data subjected to this Fourier transform,
The registered fingerprint and the collated fingerprint are collated.

The second invention (the invention according to claim 2) is the first invention.
Instead of "synthesized Fourier image data two-dimensional discrete Fourier transform and two-dimensional discrete inverse Fourier transform or subjected to one of the relative" in the invention, the amplitude suppression to a "synthesized Fourier image <br/> data it is obtained by processing the two-dimensional discrete Fourier transform after conducting and the two-dimensional discrete inverse Fourier transform or subjected to one of the "as. According to the present invention, amplitude suppression processing such as log processing and √ processing is performed on synthetic Fourier image data, and two-dimensional discrete Fourier transform or two-dimensional discrete inverse Fourier transform is performed.

The third invention (the invention according to claim 3) is the first
Instead of "subjected to two-dimensional discrete Fourier transform for the registration fingerprint image data to create the registration Fourier image data" in the invention, the amplitude suppression from two-dimensional discrete Fourier transform is performed on the image data of "registered fingerprints "Registered Fourier image data is created by performing processing". Also, instead of "by performing two-dimensional discrete Fourier transform for the collation fingerprint image data to create a collation Fourier image data", the amplitude suppression from two-dimensional discrete Fourier transform is performed on the image data of "collation fingerprint The matching Fourier image data is created by performing the processing ". According to the present invention, the two- dimensional discrete Fourier transform is applied to the image data of the registered fingerprint , and the log processing and √
Registered Fourier image data is created by performing amplitude suppression processing such as processing. Further, the two- dimensional discrete Fourier transform is applied to the image data of the collation fingerprint , and the amplitude suppressing process such as the log process or the √ process is performed, whereby the collation Fourier image data is created.

The fourth invention (the invention according to claim 4) is registered.
Create a registration Fourier image data by performing two-dimensional discrete Fourier transform on the image data of the fingerprint, the collation Fourier image by performing two-dimensional discrete Fourier transform on the image data <br/> data of the collation fingerprint
Image data is created, the registered Fourier image data and the matching Fourier image data are combined, and the frequency zero is used as the starting point for the combined Fourier image data obtained by this.
The pixel data within the range of surrounding x1 to x2 (0 <x1 <x2)
The weighting process that emphasizes the data
One alms common information suppressed rice planting two-dimensional discrete Fourier transform and two-dimensional discrete inverse Fourier transform of the fingerprint,
Individual matching the intensity registered based on the fingerprint of the correlation component finger of each data constituting the image data of the correlation component area appearing in the synthesized Fourier image data having undergone Fourier transform
The pattern is matched with the crest .

According to the present invention, the registered Fourier image data and the matching Fourier image data are synthesized, and the frequency of the synthesized Fourier image data is obtained.
A range of perimeters x1 to x2 (0 <x1 <x2) starting from zero
Weighting processing is performed to emphasize the data of the pixels in the circle,
Common information collation fingerprint and the registered fingerprint suppressed et al is a two-dimensional discrete Fourier transform or two-dimensional discrete inverse Fourier transform is performed. Then, based on the intensity of the correlation component of each individual data constituting the image data <br/> data correlation component area appearing in the synthesized Fourier image data having undergone Fourier transform, the collation fingerprint and the registered fingerprint Matching is done.

The fifth invention (the invention according to claim 5) is the fourth invention.
In the invention, " the frequency of synthetic Fourier image data is
Surroundings x1 to x2 (0 <x1 <x2) starting from wave number zero
Performs a weighting process that emphasizes the pixel data within the range
Registered fingerprint and subjected to one of the common information suppressed rice planting two-dimensional discrete Fourier transform and two-dimensional discrete inverse Fourier transform for the collation fingerprint "in place of Te," synthesized Fourier image de <br/> over data performing amplitude suppression processing for, its frequency zero for synthesized Fourier image data having undergone this amplitude suppression processing
Range of surrounding x1 to x2 (0 <x1 <x2) starting from
Register by performing weighting processing that emphasizes the data of pixels inside
One of the two- dimensional discrete Fourier transform and the two- dimensional discrete inverse Fourier transform is applied while suppressing common information between the fingerprint and the collation fingerprint . " According to the invention, combined amplitude suppression processing of log processing or √ processing on the Fourier image data is subjected to the weighting process in accordance with the frequency synthesized Fourier image data having undergone this amplitude suppression processing is performed, Two- dimensional discrete Fourier transform or two- dimensional discrete inverse Fourier transform is performed.

The sixth invention (the invention according to claim 6) is the fourth invention.
Instead of "subjected to two-dimensional discrete Fourier transform for the registration fingerprint image data to create the registration Fourier image data" in the invention, the amplitude suppression from two-dimensional discrete Fourier transform is performed on the image data of "registered fingerprints "Registered Fourier image data is created by performing processing". Also, instead of "by performing two-dimensional discrete Fourier transform for the collation fingerprint image data to create a collation Fourier image data", the amplitude suppression from two-dimensional discrete Fourier transform is performed on the image data of "collation fingerprint The matching Fourier image data is created by performing the processing ". According to the present invention, the two- dimensional discrete Fourier transform is applied to the image data of the registered fingerprint , and the log processing and √
Registered Fourier image data is created by performing amplitude suppression processing such as processing. Further, the two- dimensional discrete Fourier transform is applied to the image data of the collation fingerprint, and the amplitude suppressing process such as the log process or the √ process is performed, whereby the collation Fourier image data is created.

[0014]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, the present invention will be described in detail based on embodiments. FIG. 2 is a block configuration diagram of a fingerprint collation device showing an embodiment of the present invention. In the figure, 10 is an operation unit, 20 is a control unit, and the operation unit 10 includes a ten-key 10-1, a display (LCD).
A fingerprint sensor 10-3 is provided together with 10-2.
The fingerprint sensor 10-3 includes a light source 10-31 and a prism 10-.
32, CCD camera 10-33. The control unit 20 includes a control unit 20-1 having a CPU,
ROM 20-2, RAM 20-3, hard disk (HD) 20-4, frame memory (FM) 20-5
And an external connection unit (I / F) 20-6 and a Fourier transform unit (FFT) 20-7. The ROM 20-2 stores a registration program and a collation program.

[Registration of Fingerprint] The fingerprint of the user is registered in this fingerprint collating apparatus as follows. That is,
Before operating, the user inputs the ID number assigned to him / her by using the ten-key pad 10-1 (step 301 shown in FIG. 3), and then the prism 1 of the fingerprint sensor 10-3.
Place your finger on 0-32. Light source 1 for prism 10-32
Light is emitted from 0-31, and the prism 10-32
In the concave portion (valley line portion) of the fingerprint that does not contact the surface of the
The light from -31 is totally reflected and reaches CCD camera 10-33. On the contrary, the total reflection condition is broken at the convex portion (ridge portion) of the fingerprint that comes into contact with the surface of the prism 10-32, and the light source 10-31
Light from is scattered. As a result, a fingerprint pattern with a high contrast is obtained in which the valley portion of the fingerprint is bright and the ridge portion is dark. The pattern of this collected fingerprint (registered fingerprint) is
By the A / D conversion, it is given to the control unit 20 as a grayscale image (image data: two-dimensional pattern data) having 320 × 400 pixels and 256 gradations.

The control unit 20-1 stores the image data of the registered fingerprint given from the operation unit 10 in the frame memory 20-.
The captured image data of the registered fingerprint is reduced (step 303). This reduction processing is performed with 320 × 400 pixels, 2
Original image data of 56 gradations, its x direction (horizontal direction)
With respect to, the left and right ends are removed by 32 pixels and thinned out by 4 pixel bitches. In the y direction (vertical direction), the upper and lower ends are removed by 8 pixels and thinned out by 3 pixel pitches. As a result, the image data of the registered fingerprint is 6
The image data is reduced to 4 × 128 pixels and 256 gradations (see FIG. 5).

Then, the control unit 20-1 sends the reduced image data of the registered fingerprint (see FIG. 1A) to the Fourier transform unit 20-7, and the image data of the registered fingerprint is subjected to the two-dimensional discrete Fourier transform. Transform (DFT) is applied (step 30)
4). As a result, the image data of the registered fingerprint shown in FIG. 1A becomes Fourier image data (registered Fourier image data) as shown in FIG.

Then, the control unit 20-1 weights the Fourier image data according to the frequency (step 305), and uses the weighted Fourier image data as the original image data of the registered fingerprint, the hard disk 20. Filed in -4 corresponding to the ID number (step 306).

By the weighting process in step 305, information indicating generality of fingerprints such as patterns (common information of fingerprints) is suppressed, and information indicating individuality such as feature points (characteristic information of fingerprints) is emphasized or extracted. To do. FIG. 7 shows an example of a weighting function used in this weighting process. In this weighting function, the horizontal axis represents the pixel position (corresponding to the radius) starting from frequency zero of the Fourier image data, and the vertical axis represents the weighting coefficient by which the data at that pixel position is multiplied.

Using the weighting function shown in FIG. 7A,
Starting from the frequency zero of the Fourier image data, that is, from the center of the Fourier image shown in FIG.
The weighting coefficient “1” is multiplied to the pattern data of the surrounding x pixels. As a result, pattern data of surrounding x pixels,
That is, only low frequency components are extracted (low pass filter). If the weighting function shown in FIG. 7B is used, the frequency zero of the Fourier image data is used as the starting point, and its surroundings x1 to
The data of the pixels within the range of x2 is multiplied by the weighting factor “1”. As a result, only the pattern data within the range of the surrounding x1 to x2, that is, only the component of the predetermined frequency band is extracted (bandpass filter).

It is very difficult to determine the optimum weight function. The present applicant obtained the recognition rate for each frequency and, as a result of consideration, decided to adopt a weighting function of the type as shown in FIG. 7 (c). FIG. 8 shows the recognition rate for each frequency. With the center point of the Fourier image data set at position zero, the frequency is limited outward by the position 64 (in the x direction) by concentric circles centered at the position zero. From the position 64 to about 90, a concentric circle is formed, but it is similarly limited. The radius of the concentric circle that defines this restricted area is the horizontal axis of the graph. The vertical axis of the graph shows the value (%) obtained by obtaining the recognition rate using the information of only the area limited within the concentric circle, and the differential value per pixel unit. In addition, the area of the concentric circles (total number of pixels in the area) and the differential value for each pixel of the radius (one position at a time) are also shown.

The points to be noted in this graph are the recognition rate and its differential value. From the recognition rate, it peaks at a position of about 48 and continues to decrease thereafter. That is, there is no valid recognition information after that. As a result, the amount of information to be recognized is reduced, and the load of processing and registration memory can be reduced. Further, from the differentiation of the recognition rate, it can be read that there is particularly effective information for recognition at the positive position (radius) from about 10 to about 48. As a result, an improvement in recognition rate can be expected by weighting this area in particular. In FIG. 7C, the part on the plus side of the graph obtained by differentiating the recognition rate in FIG. 8 is of a type corresponding to the shape.

The two-dimensional discrete Fourier transform is described, for example, in "Introduction to Computer Image Processing, edited by Japan Industrial Technology Center, published by Soken Shuppan Co., Ltd., P.44-45 (Reference 1)". .

[Fingerprint Matching] In this fingerprint matching device, the user's fingerprint matching is performed as follows. That is, during operation, the user inputs the ID number assigned to him / her using the ten-key pad 10-1 (step 401 shown in FIG. 4), and then the prism 10 of the fingerprint sensor 10-3.
Put your finger on -32. As a result, the pattern of the collected fingerprint (verification fingerprint) is 3
As a grayscale image (image data: two-dimensional pattern data) of 20 × 400 pixels and 256 gradations, the control unit 2
Given to 0.

When the ID number is given via the numeric keypad 10-1, the control section 20-1 receives the hard disk 20-.
From the registered fingerprints stored in the file No. 4, Fourier image data of the registered fingerprints subjected to the weighting process corresponding to the ID number is read (step 402).

Further, the control section 20-1 stores the image data of the collation fingerprint given from the operation section 10 in the frame memory 20.
The captured image data of the collated fingerprint is subjected to the same reduction processing as that performed in step 303 (step 404). As a result, the image data of the collation fingerprint is 64 × 128 pixels,
The image data is reduced to 256 gradations.

Then, the control unit 20-1 sends the reduced collation fingerprint image data (see FIG. 1E) to the Fourier transform unit 20-7, and the two-dimensional discrete Fourier transform is applied to the collation fingerprint image data. Transform (DFT) is applied (step 40)
5). As a result, the image data of the collation fingerprint shown in FIG. 1E becomes Fourier image data (collation Fourier image data) as shown in FIG.

Then, the control unit 20-1 performs the same weighting process as that performed in step 305 on this Fourier image data (step 406), and also performs the step and the Fourier image data of the collated fingerprints subjected to this weighting process. The weighted and processed Fourier image data of the registered fingerprint read in step 402 are combined (step 407),
Obtain synthetic Fourier image data.

Here, the synthetic Fourier image data is A · B when the Fourier image data of the collated fingerprint is A · e ^j θ and the Fourier image data of the registered fingerprint is B · e ^j φ.
-It is represented by e ^{j (} θ ^- φ ⁾ . However, A, B, θ, and φ are functions of frequency (Fourier) space (u, v).

Then, A · B · e ^{j (} θ ⁻ φ ⁾ is expressed as follows: A · B · e ^{j (} θ ⁻ φ ⁾ = A · B · cos (θ−φ) + j · A · B · sin (θ− φ) ... (1), A · e ^j θ = α ₁ + jβ ₁ , B · e ^j φ =
Assuming α ₂ + jβ ₂ , A = (α ₁ ² + β ₁ ² ) ^1/2 and B =
(Α ₂ ² + β ₂ ² ) ^1/2 , θ = tan ⁻¹ (β ₁ / α ₁ ), φ =
It becomes tan ⁻¹ (β ₂ / α ₂ ). By calculating the equation (1), the synthetic Fourier image data is obtained.

Note that A · B · e ^{j (} θ ⁻ φ ⁾ = A · B · e ^j
θ ・ e ^-j φ = A ・ e ^j θ ・ B ・ e ^-j φ = (α ₁ + j
β ₁ ) · (α ₂ −jβ ₂ ) = (α ₁ · α ₂ + β ₁ ·
The combined Fourier image data may be obtained as β ₂ ) + j (α ₂ · β ₁ −α ₁ · β ₂ ).

Then, after obtaining the synthetic Fourier image data in this way, the control section 20-1 carries out an amplitude suppressing process (step 408). In this embodiment, log processing is performed as the amplitude suppression processing. That is, A · B · e ^{j (} θ ⁻ which is an arithmetic expression of the composite Fourier image data described above.
By taking the log of φ ⁾ , and letting it be log (A ・ B) ・ ej ⁽ θ ^- φ ⁾ , the amplitude A ・ B is log (A ・ B)
(A / B> log (A / B)).

FIG. 1D shows synthetic Fourier image data after the amplitude suppression processing. In the synthetic Fourier image data that has been subjected to the amplitude suppression processing, the influence of the illuminance difference between the time when the registered fingerprint is taken and the time when the collated fingerprint is taken is small. That is, by performing the amplitude suppression process, the spectrum intensity of each pixel is suppressed, the skipped value is eliminated, and more information becomes effective. In addition, by performing the amplitude suppression processing, the feature points (end points and branch points) that are personal information and the features (vortex and branch points) of the ridge are more emphasized in the fingerprint information, and the ridge line that is general fingerprint information is enhanced. The overall flow and direction can be suppressed.

In this embodiment, the log process is performed as the amplitude suppressing process, but the square process may be performed. Further, not limited to log processing and √ processing,
Any process may be used as long as the amplitude can be suppressed. If all the amplitudes are set to 1, for example, only the phase by the amplitude suppression, that is, there is an advantage that the amount of calculation can be reduced and an amount of data is reduced as compared with the log processing and the √ processing.

After performing the amplitude suppression processing in step 408, the control unit 20-1 sends the synthesized Fourier image data subjected to the amplitude suppression processing to the Fourier transform unit 20-7,
A second two-dimensional discrete Fourier transform (DFT) is performed (step 409). As a result, the synthetic Fourier image data shown in FIG. 1D becomes synthetic Fourier image data as shown in FIG.

The control unit 20-1 then proceeds to step 40.
The synthetic Fourier image data obtained in 9 is taken in, and the intensity (amplitude) of the correlation component of each pixel in a predetermined correlation component area is scanned from this synthetic Fourier image data to obtain a histogram of the intensity of the correlation component of each pixel, Upper n pixels (8 pixels in this embodiment) having a higher correlation component intensity are extracted from this histogram, and the average of the correlation component intensities of the extracted n pixels is obtained as a correlation value (score) (step 410). . Here, the correlation component area is as shown in FIG.
For the synthetic Fourier image data shown in (h), it is defined as a region S0 surrounded by a white dotted line. FIG. 6 shows a numerical example of the intensity of the correlation component of each pixel in a part of the correlation component area S0. In this figure, the values circled are the intensities of the correlation components of the top 8 pixels.

The control section 20-1 then proceeds to step 41.
The correlation value obtained in 0 is compared with a predetermined threshold value (step 411), and if the correlation value is greater than or equal to the threshold value,
It is determined that the registered fingerprint and the verification fingerprint match (step 4
12) Then, a message to that effect is displayed and the output for the electric lock is sent out. If the correlation value is less than or equal to the threshold value, it is determined that the registered fingerprint and the collated fingerprint do not match (step 413), a message to that effect is displayed, and the process returns to step 401.

Here, the threshold value to be compared with the correlation value is
As a sample, the fingerprints of the index fingers of 10 men and women in their 20s to 50s were input 10 times each, and a total of 100 fingers were used for registration and verification, respectively, and the verification was performed 10,000 times. . In this case, the correlation value at which the stranger exclusion rate is 100% is used as the threshold value. The exclusion rate of others may not be 100%, and may be set to any rate according to the purpose.

In this embodiment, from the respective pixels in the correlation component area S0, the highest n of correlation component intensity is high.
Although the pixels are extracted and the average thereof is used as the correlation value, the added value of the intensities of the correlation components of the upper n pixels may be used as the correlation value. Further, the intensities of the correlation components of all the pixels exceeding the threshold value may be added, the added value may be used as the correlation value, or the average of the added values may be used as the correlation value. Further, if at least one of the intensities of the correlation components of each pixel is equal to or more than the threshold value, it may be judged as “match”, and if the intensity exceeds the threshold value, n
Various determination methods are conceivable, such as determining “match” if there are more than one.

Further, in this embodiment, the two-dimensional discrete Fourier transform is performed in the Fourier transform unit 20-7, but it may be performed in the CPU 20-1. In addition, in this embodiment, the reduction process is performed on the image data of the registered fingerprint in step 303.
The reduction processing may be performed at a stage after reading the Fourier image data of the registered fingerprint (between steps 402 and 403). Further, the reduction process does not necessarily have to be performed on the image data of the registered fingerprint or the collation fingerprint, and the input image data may be used as it is to create the Fourier image data. If the reduction processing is performed, the capacity of the image memory used when processing the input image data can be reduced accordingly.

Further, in this embodiment, the two-dimensional discrete Fourier transform is performed in step 409 shown in FIG. 4, but the two-dimensional discrete inverse Fourier transform is performed instead of the two-dimensional discrete Fourier transform. You may do it. That is, instead of performing the two-dimensional discrete Fourier transform on the synthetic Fourier image data that has been subjected to the amplitude suppression processing,
You may make it perform a dimensional discrete inverse Fourier transform. Two
The matching accuracy of the two-dimensional discrete Fourier transform and the two-dimensional discrete inverse Fourier transform does not change quantitatively. The two-dimensional discrete inverse Fourier transform is described in the above-mentioned reference 1.

Further, in this embodiment, the amplitude suppressing process is performed in step 408 shown in FIG. 4, but the amplitude suppressing process may be omitted. By performing the amplitude suppression processing in step 408, the influence of the illuminance difference between the acquisition of the registered fingerprint and the acquisition of the collation fingerprint in the synthetic Fourier image data is reduced, and the characteristic points (end points that are personal information in the fingerprint information (end points , Bifurcation point) and ridge features (vortex, bifurcation) are more emphasized, and the matching accuracy is significantly improved.

Further, in this embodiment, amplitude suppression processing is performed on the combined Fourier image data to perform a two-dimensional discrete Fourier transform (step 40).
8, 409), the amplitude suppression processing may be performed on the Fourier image data of the registered fingerprint and the verification fingerprint before the weighting processing. That is, as shown in FIG. 9A, a step 307 for performing amplitude suppression processing is provided between steps 304 and 305 in FIG. 3, and as shown in FIG. 9B, step 408 in FIG. It may move between 405 and 406.

In this case, the amplitude suppressing process of step 307 obtains Fourier image data (registered Fourier image data) of the registered fingerprint subjected to the amplitude suppressing process as shown in FIG. By moving 408 between steps 405 and 406, FIG.
Fourier image data (verification Fourier image data) of the verification fingerprint that has been subjected to the amplitude suppression processing as shown in FIG.
Then, the Fourier image data of the registered fingerprint and the collation fingerprint subjected to the amplitude suppression processing are individually subjected to common weighting processing according to frequency, and then combined,
The synthetic Fourier image data as shown in (d) is obtained.

The suppression rate of the amplitude of the synthetic Fourier image data at this time is smaller than that in the case where the amplitude suppression processing is performed after the synthetic Fourier image data (FIG. 4). Therefore,
When the amplitude suppression processing is performed after the composite Fourier image data (FIG. 4), the matching accuracy is higher than when the amplitude suppression processing is performed and then the composite Fourier image data is used (FIG. 9). Even when the amplitude suppression processing is performed and then the synthesized Fourier image data is obtained (FIG. 9), the two-dimensional discrete inverse Fourier transform is performed on the synthesized Fourier image data instead of the two-dimensional discrete Fourier transform. May be.

Further, in this embodiment, step 30
The registered Fourier image data obtained in step 4 and the matching Fourier image data obtained in step 405 were individually subjected to common weighting processing according to frequency, but amplitude suppression processing was performed. Weighting processing may be applied to the synthetic Fourier image data. That is, FIG.
0 (a), step 305 of FIG. 3 is omitted, and as shown in FIG. 10 (b), step 406 of FIG.
May be moved between steps 408 and 409.

In this case, instead of performing the amplitude suppression processing on the combined Fourier image data, the amplitude suppression processing may be performed on the Fourier image data of the registered fingerprint and the verification fingerprint data before the combination. Good. That is, as shown in FIG. 11A, steps 304 and 3 in FIG.
06, step 307 for performing amplitude suppression processing is provided, and as shown in FIG. 11B, step 40 in FIG.
8 may be moved between steps 405 and 407. Note that, in FIG. 10, the amplitude suppression processing in step 408 may be omitted.

For reference, FIGS. 12A and 12B show respective images in the fingerprint collation process when the collated fingerprint is another person, corresponding to FIG. FIG. 1 is each image of the fingerprint matching process when the collated fingerprint is the person, and when the collated fingerprint is the person, a portion having a high correlation component intensity occurs in the correlation component area S0, but the collated fingerprint is the other person. In some cases it does not occur.

[0049]

[0050]

[0051]

As is apparent from the above description, according to the present invention, in the first, second and third inventions, the frequency zero is individually set as the starting point for the registered Fourier image data and the matching Fourier image data. Around x1 to x2 (0 <x
The same weight that emphasizes the data of pixels within the range of 1 <x2).
The registration process is performed, and the common information of the registered fingerprint and the collation fingerprint is
Suppressed , in the fourth, fifth, and sixth inventions, the composite Fourier image
Around x1 with respect to the image data starting from the frequency zero
Emphasize pixel data within x2 (0 <x1 <x2)
The registered fingerprints and collated fingerprints are shared
Information suppressed et been, by determining the weighting function in the weighting processing appropriately, the characteristic information emphasized or extracted, it is possible to improve the collation precision. In addition, the amount of calculation processing can be reduced and the matching speed can be increased.

[Brief description of drawings]

FIG. 1 is a diagram illustrating a fingerprint matching process in a fingerprint matching device according to the present invention.

FIG. 2 is a block diagram of the fingerprint collation device.

FIG. 3 is a flowchart for explaining a fingerprint registration operation in this fingerprint matching device.

FIG. 4 is a flowchart for explaining a fingerprint matching operation in this fingerprint matching device.

FIG. 5 is a diagram for explaining reduction processing for image data.

FIG. 6 is a diagram showing a numerical example of the intensity of the correlation component of each pixel in a part of the correlation component area.

FIG. 7 is a cross-sectional view illustrating a weighting function used in weighting processing.

FIG. 8 is a graph showing a recognition rate for each frequency.

FIG. 9 is a flowchart for explaining another example of the fingerprint registration operation and the fingerprint collation operation.

FIG. 10 is a flowchart for explaining another example of the fingerprint registration operation and the fingerprint collation operation.

FIG. 11 is a flowchart for explaining another example of the fingerprint registration operation and the fingerprint collation operation.

FIG. 12 is a diagram showing images corresponding to FIG. 1 in a fingerprint matching process when the matching fingerprint is another person.

[Explanation of symbols]

10 ... Operation part, 20 ... Control part, 10-1 ... Numeric keypad, 10-2 ... Display (LCD), 10-3 ...
Fingerprint sensor, 10-31 ... Light source, 10-32, Prism, 10-33 ... CCD camera, 20-1 ... Control unit, 2
0-2 ... ROM, 20-3 ... RAM, 20-4 ... Hard disk (HD), 20-5 ... Frame memory (F
M), 20-6 ... External connection section (I / F), 20-7 ... Fourier transform section (FFT).

   ─────────────────────────────────────────────────── ─── Continued front page       (56) References Japanese Patent Laid-Open No. 5-159056 (JP, A)                 JP-A-2-156387 (JP, A)                 JP-A-3-191482 (JP, A)                 JP-A-4-326680 (JP, A)                 JP-A-6-259564 (JP, A)

Claims (6)

(57) [Claims] 1. A performs the registration Fourier image data generating means for generating registration Fourier image data by performing two-dimensional discrete Fourier transform on the image data of the registration fingerprint, the two-dimensional discrete Fourier transform for the collation fingerprint image data Matching Fourier image Creating a matching Fourier image
Image data creating means, the registered Fourier image data and the matching Fourier image
Surrounding image data individually starting from the frequency zero
Data of pixels in the range of x1 to x2 (0 <x1 <x2)
The same weighting process that emphasizes
Weighting processing means for suppressing the common information of the collation fingerprints, the registered Fourier image data weighted by the weighting processing means and the matching Fourier image data are synthesized, and 2 is obtained for the synthesized Fourier image data obtained thereby. and either the performing data processing unit dimension discrete Fourier transform and two-dimensional discrete inverse Fourier transform, the image of the correlation component area appearing by the data processing means in synthesized Fourier image data having undergone Fourier transform <br / > A fingerprint collation device comprising: fingerprint collation means for collating the registered fingerprint with the collation fingerprint based on the intensity of the correlation component of each individual piece of data. 2. A subjected registration Fourier image data generating means for generating registration Fourier image data by performing two-dimensional discrete Fourier transform on the image data of the registration fingerprint, the two-dimensional discrete Fourier transform for the collation fingerprint image data Matching Fourier image Creating a matching Fourier image
Image data creating means, the registered Fourier image data and the matching Fourier image
Surrounding image data individually starting from the frequency zero
Data of pixels in the range of x1 to x2 (0 <x1 <x2)
The same weighting process that emphasizes
Weighting processing means for suppressing the common information of the collation fingerprints, the registered Fourier image data weighted by the weighting processing means and the collation Fourier image data are synthesized, and the amplitude of the synthesized Fourier image data obtained thereby is combined. and either the performing data processing means suppression processing two-dimensional discrete Fourier transform after conducting and the two-dimensional discrete inverse Fourier transform, correlation appearing by the data processing means in synthesized Fourier image data having undergone Fourier transform A fingerprint collation unit for collating the registered fingerprint with the collation fingerprint based on the intensity of the correlation component for each individual data forming the image data of the component area. Fingerprint matching device. 3. A registration Fourier image de <br/> over data generating means for generating registration Fourier image data by from two-dimensional discrete Fourier transform is performed on the image data of the registration fingerprint by performing amplitude suppression processing, collation A two- dimensional discrete Fourier transform is applied to the image data of the fingerprint , and then amplitude suppression processing is performed to obtain the matching Fourier image.
And collation Fourier image data creating means for creating image data, the registration Fourier image data and the collation Fourier image
Surrounding image data individually starting from the frequency zero
Data of pixels in the range of x1 to x2 (0 <x1 <x2)
The same weighting process that emphasizes
Weighting processing means for suppressing the common information of the collation fingerprints, the registered Fourier image data weighted by the weighting processing means and the matching Fourier image data are synthesized, and 2 is obtained for the synthesized Fourier image data obtained thereby. and either the performing data processing unit dimension discrete Fourier transform and two-dimensional discrete inverse Fourier transform, the image of the correlation component area appearing by the data processing means in synthesized Fourier image data having undergone Fourier transform <br / > A fingerprint collation device comprising: fingerprint collation means for collating the registered fingerprint with the collation fingerprint based on the intensity of the correlation component of each individual piece of data. 4. A subjected registration Fourier image data generating means for generating registration Fourier image data by performing two-dimensional discrete Fourier transform on the image data of the registration fingerprint, the two-dimensional discrete Fourier transform for the collation fingerprint image data Matching Fourier image Creating a matching Fourier image
An image data creating means, the registration Fourier image data and synthesizes the said collation Fourier image de <br/> over data, thereby synthesized Fourier image obtained
Around x1 with respect to the image data starting from the frequency zero
Emphasize pixel data within x2 (0 <x1 <x2)
The verification fingerprint and the registered fingerprint weighted handle the going to be
Common information suppressed rice planting two-dimensional discrete Fourier transform and two-dimensional one to apply Day inverse discrete Fourier transform of
The registration based and data processing means, the strength of the correlation components of the respective individual data constituting the image <br/> data correlation component area appearing in the synthesized Fourier image data having undergone Fourier transform by the data processing means fingerprint collation apparatus characterized by comprising a fingerprint <br/> matching means for matching the fingerprint and the collation fingerprint. 5. subjecting the registration Fourier image data generating means for generating registration Fourier image data by performing two-dimensional discrete Fourier transform on the image data of the registration fingerprint, the two-dimensional discrete Fourier transform for the collation fingerprint image data Matching Fourier image Creating a matching Fourier image
An image data creating means, the registration Fourier image data and synthesizes the said collation Fourier image de <br/> over data, thereby synthesized Fourier image obtained
Performing amplitude suppression processing for the image data, the frequency for the synthesized Fourier image data having undergone this amplitude suppression processing
Of surrounding x1-x2 (0 <x1 <x2) starting from the number zero
Perform weighting processing to emphasize the data of the pixels within the range
In addition to suppressing common information between the registered fingerprint and the collation fingerprint, 2
And either the performing data processing unit dimension discrete Fourier transform and two-dimensional discrete inverse Fourier transform, the image of the correlation component area appearing by the data processing means in synthesized Fourier image data having undergone Fourier transform <br / > A fingerprint collation device comprising: fingerprint collation means for collating the registered fingerprint with the collation fingerprint based on the intensity of the correlation component of each individual piece of data. 6. A registration Fourier image de <br/> over data generating means for generating registration Fourier image data by from two-dimensional discrete Fourier transform is performed on the image data of the registration fingerprint by performing amplitude suppression processing, collation A two- dimensional discrete Fourier transform is applied to the image data of the fingerprint , and then amplitude suppression processing is performed to obtain the matching Fourier image.
And collation Fourier image data creating means for creating image data, the registration Fourier image data and synthesizes the said collation Fourier image de <br/> over data, thereby synthesized Fourier image obtained
Around the origin the frequency zero for the image data x1~
Emphasize pixel data within x2 (0 <x1 <x2)
The registered fingerprint and the collation fingerprint are weighted by
Common information suppressed rice planting two-dimensional discrete Fourier transform and two-dimensional one to apply Day inverse discrete Fourier transform of
The registration based and data processing means, the strength of the correlation components of the respective individual data constituting the image <br/> data correlation component area appearing in the synthesized Fourier image data having undergone Fourier transform by the data processing means fingerprint collation apparatus characterized by comprising a fingerprint <br/> matching means for matching the fingerprint and the collation fingerprint.