Fingerprint verification method

• US 5,239,590 A
• Filed: 08/05/1991
• Issued: 08/24/1993
• Est. Priority Date: 08/07/1990
• Status: Expired due to Term

• Current Assignee: Sharp Corporation (Hon Hai Precision Industry Co., Ltd.)
• Original Assignee: YOZAN, Inc.
• Inventors: Yamamoto, Makoto
• Primary Examiner(s): Mancuso, Joseph
• Application Number:
  US07/740,603
• Time in Patent Office: 750 Days
• Field of Search: 382/4, 382/5, 382/18, 382/30, 382/34, 382/42, 356/71, 340/825.34
• US Class Current: 382/125
• CPC Class Codes: G06V 40/1347 Preprocessing; Feature extraction
• CPC Class Codes: G07C 9/257 electronically G07C9/26 takes precedence

Abstract

A method for verifying that a sample image of a fingerprint is that of a designated person through comparison of sample data, generated from the sample image, with master data, generated from a master image of the designated person'"'"'s fingerprint. The master and sample image have been divided into a plurality of blocks, and each block has been divided into a plurality of block areas. Each block area has a number of pixels each having an associated direction as the sample and master data. The direction associated with each pixel was determined based on pixel density partial differentials between the pixel and adjacent pixels for a plurality of directions. The direction of the minimum pixel density partial differential for a pixel is chosen as the direction for that pixel. Based on the sample and master data fingerprint verification is determined by at least one of total dispersion, cross-correlation, and distance between classes.

Claims

1. A method for verifying that a sample image of a fingerprint is that of a designated person through comparison of sample data, generated from the sample image, with master data, generated from a master image of the designated person'"'"'s fingerprint, the master and sample image having been divided into a plurality of blocks and each block having been divided into a plurality of block areas, each block area is comprised of pixels each having an associated direction as the sample and master data, the method comprising the steps of:

determining a dispersion within a class of each block of the sample image based on the sample data which represents a direction of a finger characteristic within a block;

determining a dispersion within a class of each block of the master image based on the master data which represents a direction of a finger characteristic within a block;

determining the aberration of direction between blocks of the sample image and master image representing a difference in a finger characteristic between respective blocks of the sample image and master image based on the sample data and master data;

determining a total dispersion based on the dispersion within a class for each block of the sample image and the master image and the aberration of direction which represents a measure of the similarity between the sample image and the master image;

comparing the total dispersion for each block to a reference total dispersion; and

verifying a fingerprint as that of the designated person based on results from the comparisons.

2. The method as in claim 1, wherein the steps of determining the dispersions within a class for each block of the sample image and master image comprises the steps of:

determining an aberration of direction between a plurality of direction codes and a direction of a pixel for each pixel in each block area which represents a difference in direction between a direction of a pixel and a direction code;

determining a dispersion within a class of each block area for each direction code, for each direction code the determination of a dispersion within a class of each block area comprises the steps of,

multiplying each pixel within a block area by a weight to form a pixel product, the weight corresponding to the aberration of direction between the pixel and the direction code;

summing the pixel products of the pixels in the block area;

dividing the summed pixel products for the block area by the number of pixels in the block area to determine a dispersion within a class of the block area for that direction code;

determining as the dispersion within a class of a block area a one of the dispersions within a class of the block area for each direction code which is a minimum;

summing as a block dispersion sum, the dispersions within a class of a block area for a block; and

dividing the block sum of a block by a number of block areas in the block to determine the dispersion within a class of the block.

3. The method as in claim 2, wherein the step of determining the aberration of direction comprises the steps of:

providing for each block area of the sample and master image a direction associated with a fingerprint characteristic of a block area, the provided direction being the direction code for which the dispersion within a class of the block area for the direction codes was a minimum;

determining a direction, which is associated with a fingerprint characteristic of a block, for each block in the sample and master image by calculating the mean of the provided directions of each block area within a respective block;

determining an aberration of direction between corresponding blocks of the sample and master image based on the determined directions of the blocks in the sample and master image.

4. The method of claim 1, wherein the step of determining the aberration of direction comprises the steps of:

determining a direction, which is associated with a fingerprint characteristic of a block area, for each block area in a block for both the sample and master image based on the sample and master data;

determining a direction, which is associated with a fingerprint characteristic of a block, for each block in the sample and master image by calculating the mean of the directions of each block area within a respective block;

determining an aberration of direction between corresponding blocks of the sample and master image based on the determined directions of the blocks in the sample and master image.

5. A method for verifying that a sample image of a fingerprint is that of a designated person through comparison of sample data, generated from the sample image, with master data, generated from a master image of the designated person'"'"'s fingerprint, the master and sample image having been divided into a plurality of blocks and each block having been divided into a plurality of block areas, each block area is comprised of pixels each having an associated direction as the sample and master data, the method comprising the steps of:

determining a dispersion within a class of each block of the sample image based on the sample data which represents a direction of a finger characteristic within a block;

determining a dispersion within a class of each block of the master image based on the master data which represents a direction of a finger characteristic within a block;

determining the aberration of direction between blocks of the sample image and master image representing a difference in a finger characteristic between respective blocks of the sample image and master image based on the sample data and master data;

determining a total dispersion based on the dispersion within a class for each block of the sample image and the master image and the aberration of direction which represents a measure of the similarity between the sample image and the master image;

determining a cross-correlation between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a similarity in directions associated with the pixel sin a block area;

calculating a mean of the cross-correlations between the block areas for each block of the sample and master image to determine a cross-correlation between each block of the sample and master image;

first comparing the cross-correlations between each block of the sample and master image to a reference cross-correlation;

second comparing the total dispersion for each block to a reference total dispersion; and

verifying a fingerprint as that of the designated person based on results from the first and second comparisons.

6. A method for verifying that a sample image of a fingerprint is that of a designated person through comparison of sample data, generated from the sample image, with master data, generated from a master image of the designated person'"'"'s fingerprint, the master and sample image having been divided into a plurality of blocks and each block having been divided into a plurality of block areas, each block area is comprised of pixels each having an associated direction as the sample and master data, the method comprising the steps of:

determining a dispersion within a class of each block of the sample image based on the sample data which represents a direction of a finger characteristic within a block;

determining a dispersion within a class of each block of the master image based on the master data which represents a direction of a finger characteristic within a block;

determining the aberration of direction between blocks of the sample image and master image representing a difference in a finger characteristic between respective blocks of the sample image and master image based on the sample data and master data;

determining a total dispersion based on the dispersion within a class for each block of the sample image and the master image and the aberrations of direction which represents a measure of the similarity between the sample image and the master image;

determining a distance between classes between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a difference in directions associated with the pixels in a block area;

calculating a mean of the distance between classes between the block areas for each block of the sample and master data to determine a distance between classes between each block of the sample and master data;

first comparing the distance between classes between each block of the sample and master data to a reference distance between classes;

second comparing the total dispersion for each block to a reference total dispersion; and

verifying a fingerprint as that of the designated person based on results from the first and second comparisons.

7. A method for generating sample and master data for fingerprint verification, the method comprising the steps of:

inputting a gray scale image, which is one of a master image or a sample image, of a finger including a tip of said finger using an image processing portion, said gray scale image composed of pixels;

outputting the gray scale image from said image processing portion to a data processing portion;

determining a center of characteristics of the gray scale image using the data processing portion;

taking as an area of examination that part of the gray scale image from the center of characteristics to said tip of said finger using the data processing portion;

calculating for each pixel in the area of examination density partial differentials between said pixel and at least adjacent pixels in a plurality of directions using the data processing portion; and

determining the master data or sample data in the data processing portion, the master data or sample data being a direction of each pixel in the area of examination of the master image or sample image, respectively, the direction of a pixel being a minimum density partial differential for each pixel in the area of examination.

8. The method of claim 7, wherein after the step of taking, the method comprises the additional steps of:

dividing the area of examination into a plurality of blocks in the image processing portion; and

determining a direction of a block for the master data or the sample data in the image processing portion, the direction of a block being the direction of a majority of the pixels in the block.

9. The method of claim 7, wherein after the step of taking, the method comprises the additional steps of:

dividing the area of examination into a plurality of blocks in the image processing portion;

dividing each block into a plurality of block areas in the image processing portions; and

determining a direction of a block area for the master data or the sample data in the image processing portion, the direction of a block area being the direction of a majority of the pixels in the block area.

10. The method as in claim 7, wherein the step of determining the center of characteristics comprises the steps of:

determining partial differentials of density values between adjacent pixel sin both vertical and horizontal directions of the gray scale image;

making histograms of the determined partial differentials of density values in the vertical and horizontal directions; and

determining the center of characteristics of the gray scale image based on the histograms.

11. The method as in claim 7, wherein the step of determining the center of characteristics comprises the steps of:

taking a partial differential of adjacent pixels in the gray scale image in both a vertical, y coordinate axis, direction and horizontal, x coordinate axis, direction,

summing the partial differentials between pixels parallel to the y-axis for each coordinate of the x-axis,

determining as an x-max coordinate, a coordinate on the x-axis corresponding to a sum of partial differentials parallel to the y-axis which is a maximum,

summing the partial differentials between pixels parallel to the x-axis for each coordinate of the y-axis,

determining as a y-max coordinate, a coordinate on the y-axis corresponding to a sum of partial differentials parallel to the x-axis which is a maximum, and

designating as the center of characteristics for the gray scale image the x-max and y-max coordinates.

12. A method for generating sample and master data for fingerprint verification, the method comprising the steps of:

inputting a gray scale image, which is one of a master image or a sample image, of a finger including a tip of said finger using an image processing portion, said gray scale image composed of pixels;

outputting the gray scale image from said image processing portion to a data processing portion;

determining a center of characteristics for the gray scale image using the data processing portion;

taking as an area of examination an area having said center of characteristics as its center using the data processing portion;

calculating for each pixel in the area of examination density partial differentials between said pixel and at least adjacent pixels in a plurality of directions using the data processing portion; and

determining the master data or sample data in the data processing portion, the master data or sample data being a direction of each pixel in the area of examination of the master image or sample image, the direction of a pixel being a minimum density partial differential for each pixel in the area of examination.

13. The method of claim 12, wherein after the step of taking, the method comprises the additional steps of:

dividing the area of examination into a plurality of blocks in the image processing portions; and

determining a direction of a block for the master data or the sample data in the image processing portion, the direction of a block being the direction of a majority of the pixels in the block.

14. The method of claim 12, wherein after the step of taking, the method comprises the additional steps of:

dividing the area of examination into a plurality of blocks in the image processing portion;

dividing each block into a plurality of block areas in the image processing portion; and

determining a direction of a block area for the master data or the sample data in the image processing portion, the direction of a block area being the direction of a majority of the pixels in the block area.

15. The method as in claim 12, wherein the step of determining the center of characteristics comprises the steps of:

determining partial differentials of density values between adjacent pixels in both vertical and horizontal directions of the gray scale image;

making histograms of the determined partial differentials of density values in the vertical and horizontal directions; and

determining the center of characteristics of the gray scale image based on the histograms.

16. The method as in claim 12, wherein the step of determining the center of characteristics comprises the steps of:

taking a partial differential of adjacent pixels in the gray scale image in both a vertical, y coordinate axis, direction and horizontal, x coordinate axis, direction,

summing the partial differentials between pixels parallel to the y-axis for each coordinate of the x-axis,

determining as an x-max coordinate, a coordinate on the x-axis corresponding to a sum of partial differentials parallel to the y-axis which is a maximum,

summing the partial differentials between pixels parallel to the x-axis for each coordinate of the y-axis,

determining as a y-max coordinate, a coordinate on the y-axis corresponding to a sum of partial differentials parallel to the x-axis which is a maximum, and

designating as the center of characteristics for the gray scale image the x-max and y-max coordinates.

17. A method for generating sample and master data for fingerprint verification, the method comprising the steps of:

inputting a gray scale image, which is one of a master image or a sample image, of a finger including a fillet center of said finger using an image processing portions, said gray scale image composed of pixels;

outputting the gray scale image from said image processing portion to a data processing portion;

taking as an area of examination an area having said center of characteristics as its center and including said fillet center using the data processing portion;

calculating for each pixel in the area of examination density partial differentials between said pixel and at least adjacent pixels in a plurality of directions using the data processing portions; and

determining the master data or sample data in the data processing portion, the master data or sample data being a direction of each pixel in the area of examination of the master image or sample image, the direction of a pixel being a minimum density partial differential for each pixel in the area of examination.

18. The method of claim 17, wherein after the step of taking, the method comprises the additional steps of:

dividing the area of examination into a plurality of blocks in the image processing portion; and

determining a direction of a block for the master data or the sample data in the image processing portion, the direction of a block being the direction of a majority of the pixels in the block.

19. The method of claim 17, wherein after the step of taking, the method comprises the additional steps of:

dividing the area of examination into a plurality of blocks in the image processing portion;

dividing each block into a plurality of block areas in the image processing portion; and

determining a direction of a block area for the master data or the sample data in the image processing portion, the direction of a block area being the direction of a majority of the pixels in the block area.

20. The method as in claim 17, wherein the step of determining the center of characteristics comprises the steps of:

determining partial differentials of density values between adjacent pixels in both vertical and horizontal directions of the gray scale image;

making histograms of the determined partial differentials of density values in the vertical and horizontal directions; and

determining the center of characteristics of the gray scale image based on the histograms.

21. The method as in claim 17, wherein the step of determining the center of characteristics comprises the steps of:

taking a partial differential of adjacent pixels in the gray scale image in both a vertical, y coordinate axis, direction and horizontal, x coordinate axis, direction,

summing the partial differentials between pixels parallel to the y-axis for each coordinate of the x-axis,

determining as an x-max coordinate, a coordinate on the x-axis corresponding to a sum of partial differentials parallel to the y-axis which is a maximum,

summing the partial differentials between pixels parallel to the x-axis for each coordinate of the y-axis,

determining as a y-max coordinate, a coordinate on the y-axis corresponding to a sum of partial differentials parallel to the x-axis which is a maximum, and

designating as the center of characteristics for the gray scale image the x-max andy-max coordinates.

22. A method for verifying that a sample image of a fingerprint is that of a designated person through comparison of sample data, generated from the sample image, with master data, generated from a master image of the designated person'"'"'s fingerprint, the method comprising the steps of:

inputting a gray scale image, which is one of the master image or the sample image, of a finger including a tip of said finger using an image processing portion, said gray scale image composed of pixels;

outputting the gray scale image from said image processing portion to a data processing portion;

determining a center of characteristics of the gray scale image using the data processing portion;

taking as an area of examination that part of the gray scale image from the center of characteristics to said tip of said finger using the data processing portion;

calculating density partial differentials for each pixel in the area of examination between said pixel and at least adjacent pixels in a plurality of directions using the data processing portion;

determining the master data or the sample data in the data processing portion, the master data or sample data being a direction of each pixel in the area of examination of the master image or sample image, respectively, the direction of a pixel being a minimum density partial differential for each pixel in the area of examination; and

performing fingerprint verification based on the sample and master data in the data processing portion.

23. The method of claim 22, wherein after the step of taking, the method comprises the additional steps of:

dividing the area of examination into a plurality of blocks in the image processing portion; and

determining a direction of a block for the master data or the sample data in the image processing portion, the direction of a block being the direction of a majority of the pixels in the block.

24. The method of claim 22, wherein after the step of taking, the method comprises the additional steps of:

dividing the area of examination into a plurality of blocks in the image processing portion;

dividing each block into a plurality of block areas in the image processing portion; and

determining a direction of a block area for the master data or the sample data in the image processing portion, the direction of a block area being the direction of a majority of the pixels in the block area.

25. The method as in claim 22, wherein the step of determining the center of characteristics comprises the steps of:

determining partial differentials of density values between adjacent pixels in both vertical and horizontal directions of the gray scale image;

making histograms of the determined partial differentials of density values in the vertical and horizontal directions; and

determining the center of characteristics of the gray scale imaged based on the histograms.

26. The method as in claim 22, wherein the step of determining the center of characteristics comprises the steps of:

taking a partial differential of adjacent pixels in the gray scale image in both a vertical, y coordinate axis, direction and horizontal, x coordinate axis, direction,

summing the partial differentials between pixels parallel to the y-axis for each coordinate of the x-axis,

determining as an x-max coordinate, a coordinate on the x-axis corresponding to a sum of parallel differentials parallel to the y-axis which is a maximum,

summing the partial differentials between pixels parallel to the x-axis for each coordinate of the y-axis,

determining as a y-max coordinate, a coordinate on the y-axis corresponding to a sum of partial differentials parallel to the x-axis which is a maximum, and

designating as the center of characteristics for the gray scale image the x-max and y-max coordinates.

27. The method of claim 22, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block areas;

determining a dispersion within a class of each block of the sample image based on the sample data which represents a direction of a finger characteristic within a block;

determining a dispersion within a class of each block of the master image based on the master data which represents a direction of a finger characteristic within a block;

determining the aberration of direction between blocks of the sample image and master image representing a difference in a finger characteristic between respective blocks of the sample image and master image based on the sample data and master data;

determining a total dispersion based on the dispersion within a class for each block of the sample image and the master image and the aberration of direction which represents a measure of the similarity between the sample image and the master image;

comparing the total dispersion for each block to a reference total dispersion; and

verifying a fingerprint as that of the designated person based on results from the comparisons.

28. The method as in claim 27, wherein the steps of determining the dispersions within a class for each block of the sample image and master image comprises the steps of:

determining an aberration of direction between a plurality of direction codes and a direction of a pixel for each pixel in each block area which represents a difference in direction between a direction of a pixel and a direction code;

determining a dispersion within a class of each block area for each direction code, for each direction code the determination of a dispersion within a class of each block area comprises the steps of,

multiplying each pixel within a block area by a weight to form a pixel product, the weight corresponding to the aberration of direction between the pixel and the direction code;

summing the pixel products of the pixels in the block area;

dividing the summed pixel products for the block area by the number of pixels in the block area to determine a dispersion within a class of the block area for that direction code;

determining as the dispersion within a class of a block area a one of the dispersions within a class of the block area for each direction code which is a minimum;

summing as a block dispersion sum, the dispersions within a class of a block area for a block; and

dividing the block sum of a block by a number of block areas in the block to determine the dispersion within a class of the block.

29. The method as in claim 28, wherein the step of determining the aberration of direction comprises the steps of:

providing for each block area of the sample and master image a direction associated with a fingerprint characteristic of a block area, the provided direction being the direction code for which the dispersion within a class of the block area for the direction codes was a minimum;

determining a direction, which is associated with a fingerprint characteristic of a block, for each block in the sample and master image by calculating the mean of the provided directions of each block area within a respective block;

determining an aberration of direction between corresponding blocks of the sample and master image based on the determined directions of the blocks in the sample and master image.

30. The method as in claim 27, wherein the step of determining the aberration of direction comprises the steps of:

determining a direction, which is associated with a fingerprint characteristic of a block area, for each block area in a block for both the sample and master image based on the sample and master data;

determining a direction, which is associated with a fingerprint characteristic of a block, for each block in the sample and master image by calculating the mean of the directions of each block area within a respective block;

determining an aberration of direction between corresponding blocks of the sample and master image based on the determined directions of the blocks in the sample and master image.

31. The method of claim 22, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block areas;

determining a cross-correlation between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a similarity in directions associated with the pixels in a block area;

calculating a mean of the cross-correlations between the block areas for each block of the sample and master image to determine a cross-correlation between each block of the sample and master image;

comparing the cross-correlations between each block of the sample and master image to a reference cross-correlation; and

verifying a fingerprint as that of the designated person based on results from the comparisons.

32. The method of claim 22, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block areas;

determining a distance between classes between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a difference in directions associated with the pixels in a block area;

calculating a mean of the distance between classes between the block areas for each block of the sample and master data to determine a distance between classes between each block of the sample and master data;

comparing the distance between classes between each block of the sample and master data to a reference distance between classes; and

verifying a fingerprint as that of the designated person based on results from the comparisons.

33. The method of claim 22, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block areas;

determining a dispersion within a class of each block of the sample image based on the sample data which represents a direction of a finger characteristic within a block;

determining a dispersion within a class of each block of the master image based on the master data which represents a direction of a finger characteristic within a block;

determining the aberration of direction between blocks of the sample image and master image representing a difference in a finger characteristic between respective blocks of the sample image and master image based on the sample data and master data;

determining a total dispersion based on the dispersion within a class for each block of the sample image and the master image and the aberration of direction which represents a measure of the similarity between the sample image and the master image;

determining a cross-correlation between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a similarity in directions associated with the pixels in a block area;

calculating a mean of the cross-correlations between the block areas for each block of the sample and master image to determine a cross-correlation between each block of the sample and master image;

comparing the cross-correlations between each block of the sample and master image to a reference cross-correlation;

second comparing the total dispersion for each block to a reference total dispersion; and

verifying a fingerprint as that of the designated person based on results from the first and second comparisons.

34. The method of claim 22, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block areas;

determining a dispersion within a class of each block of the sample image based on the sample data which represents a direction of a finger characteristic within a block;

determining a dispersion within a class of each block of the master image based on the master data which represents a direction of a finger characteristic within a block;

determining the aberration of direction between blocks of the sample image and master image representing a difference in a finger characteristic between respective blocks of the sample image and master image based on the sample data and master data;

determining a total dispersion based on the dispersion within a class for each block of the sample image and the master image and the aberration of direction which represents a measure of the similarity between the sample image and the master image;

determining a distance between classes between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a difference in directions associated with the pixels in a block area;

calculating a mean of the distance between classes between the block areas for each block of the sample and master data to determine a distance between classes between each block of the sample and master data;

first comparing the distance between classes between each block of the sample and master data to a reference distance between classes;

second comparing the total dispersion for each block to a reference total dispersion; and

verifying a fingerprint as that of the designated person based on results from the first and second comparisons.

35. The method of claim 22, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block areas;

determining a cross-correlation between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a similarity in directions associated with the pixels in a block area;

calculating a mean of the cross-correlations between the block areas for each block of the sample and master image to determine a cross-correlation between each block of the sample and master image;

determining a distance between classes between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a difference in directions associated with the pixels in a block area;

calculating a mean of the distance between classes between the block areas for each block of the sample and master data to determine a distance between classes between each block of the sample and master data;

first comparing the distance between classes between each block of the sample and master data to a reference distance between classes;

second comparing the cross-correlations between each block of the sample and master image to a reference cross-correlation;

verifying a fingerprint as that of the designated person based on results from the first and second comparisons.

36. The method of claim 22, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of blocks areas;

determining a dispersion within a class of each block of the sample image based on the sample data which represents a direction of a finger characteristic within a block;

determining a dispersion within a class of each block of the master image based on the master data which represents a direction of a finger characteristic within a block;

determining the aberration of direction between blocks of the sample image and master image representing a difference in finger characteristic between respective blocks of the sample image and master image based on the sample data and master data;

determining a total dispersion based on the dispersion within a class for each block of the sample image and the master image and the aberration of direction which represents a measure of the similarity between the sample image and the master image;

determining a cross-correlation between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a similarity in directions associated with the pixels in a block area;

calculating a mean of the cross-correlations between the block areas for each block of the sample and master image to determine a cross-correlation between each block of the sample and master image;

determining a distance between classes between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a difference in directions associated with the pixels in a block area;

calculating a mean of the distance between classes between the block areas for each block of the sample and master data to determine a distance between classes between each block of the sample and master data;

first comparing the distance between classes between each block of the sample and master data to a reference distance between classes;

second comparing the cross-correlations between each block of the sample and master image to a reference cross-correlation;

third comparing the total dispersion for each block to a reference total dispersion; and

verifying a fingerprint as that of the designated person based on results from the first, second and third comparisons.

37. A method for generating sample and master data for fingerprint verification, the method comprising the steps of:

inputting a gray scale image, which is one of the master image or the sample image, of a finger including a tip of said finger, using an image processing portion, said gray scale image composed of pixels;

outputting the gray scale image from said image processing portion to a data processing portion;

determining a center of characteristics for the gray scale image using the data processing portion;

taking as an area of examination an area having said center of characteristics as its center using the data processing portion;

calculating density partial differentials for each pixel in the area of examination between said pixel and at least adjacent pixels in a plurality of directions using the data processing portion;

determining the master data or the sample data in the data processing portion, the master data or sample data being a direction of each pixel in the area of examination of the master image or sample image, respectively, the direction of a pixel being a minimum density partial differential for each pixel in the area of examination; and

performing fingerprint verification based on the sample and master data in the data processing portion.

38. The method of claim 37, wherein after the step of taking, the method comprises the additional steps of:

dividing the area of examination into a plurality of blocks in the image processing portion; and

determining a direction of a block for the master data for the sample data in the image processing portion, the direction of a block being the direction of a majority of the pixels in the block.

39. The method of claim 37, wherein after the step of taking, the method comprises the additional steps of:

dividing the area of examination into a plurality of blocks in the image processing portion;

dividing each block into a plurality of block areas in the image processing portion; and

determining a direction of a block area for the master data or the sample data in the image processing portion, the direction of a block area being the direction of a majority of the pixels in the block area.

40. The method as in claim 37, wherein the step of determining the center of characteristics comprises the steps of:

determining partial differentials of the density values between adjacent pixels in both vertical and horizontal directions of the gray scale image;

making histograms of the determined partial differentials of density values in the vertical and horizontal directions; and

determining the center of characteristics of the gray scale image based on the histograms.

41. The method as in claim 37, wherein the step of determining the center of characteristics comprises the steps of:

taking a partial differential of adjacent pixels in the gray scale image in both a vertical, y coordinate axis, direction and horizontal, x coordinate axis, direction,

summing the partial differentials between pixels parallel to the y-axis for each coordinate of the x-axis,

determining as an x-max coordinate, a coordinate on the x-axis corresponding to a sum of partial differential parallel to the y-axis which is a maximum,

summing the partial differentials between pixels parallel to the x-axis for each coordinate of the y-axis,

determining as a y-max coordinate, a coordinate on the y-axis corresponding to a sum of partial differentials parallel to the x-axis which is a maximum, and

designating as the center of characteristics for the gray scale image the x-max and y-max coordinates.

42. The method of claim 37, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block areas;

determining a dispersion within a class of each block of the sample image based on the sample data which represents a direction of a finger characteristic within a block;

determining a dispersion within a class of each block of the master image based on the master data which represents a direction of a finger characteristic within a block;

determining the aberration of direction between blocks of the sample image and master image representing a difference in a finger characteristic between respective blocks of the sample image and master image based on the sample data and master data;

determining a total dispersion based on the dispersion within a class for each block of the sample image and the master image and the aberration of direction which represents a measure of the similarity between the sample image and the master image;

comparing the total dispersion for each block to a reference total dispersion; and

verifying a fingerprint as that of the designated person based on result from the comparisons.

43. The method as in claim 42, wherein the steps of determining the dispersion within a class for each block of the sample image and master image comprises the steps of:

determining an aberration of direction between a plurality of direction codes and a direction of a pixel for each pixel in each block area which represents a difference in direction between a direction of a pixel and a direction code;

determining a dispersion within a class of each block area for each direction code, for each direction code the determination of a dispersion within a class of each block area comprises the steps of,

multiplying each pixel within a block area by a weight to form a pixel product, the weight corresponding to the aberration of direction between the pixel and the direction code;

summing the pixel products of the pixels in the block area;

dividing the summed pixel products for the block area by the number of pixels in the block area to determine a dispersion within a class of the block area for that direction code;

determining as the dispersion within a class of a block area a one of the dispersions within a class of the block area for each direction code which is a minimum;

summing as a block dispersion sum, the dispersion within a class of a block area for a block; and

dividing the block sum of a block by a number of block areas in the block to determine the dispersion within a class of the block.

44. The method as in claim 43, wherein the step of determining the aberration of direction comprises the steps of:

providing for each block area of the sample and master image a direction associated with a fingerprint characteristic of a block area, the provided direction being the direction code for which the dispersion within a class of the block area for the direction codes was a minimum;

determining a direction, which is associated with a fingerprint characteristic of a block, for each block in the sample and master image by calculating the mean of the provided directions of each block area within a respective block;

determining an aberration of direction between corresponding blocks of the sample and master image based on the determined directions of the blocks in the sample and master image.

45. The method as in claim 42, wherein the step of determining the aberration of a direction comprises the steps of:

determining a direction, which is associated with a fingerprint characteristic of a block area, for each block area in a block for both the sample and master image based on the sample and master data;

determining a direction, which is associated with a fingerprint characteristic of a block, for each block in the sample and master image by calculating the mean of the directions of each block area within a respective block;

determining an aberration of direction between corresponding blocks of the sample and master image based on the determined directions of the blocks in the sample and master image.

46. The method of claim 37, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block areas;

determining a cross-correlation between the blocks areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a similarity in directions associated with the pixels in a block area;

calculating a mean of the cross-correlations between the block areas for each block of the sample and master image to determine a cross-correlation between each block of the sample and master image;

comparing the cross-correlations between each block of the sample and master image to a reference cross-correlation; and

verifying a fingerprint as that of the designated person based on results from the comparisons.

47. The method of claim 37, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block areas;

determining a distance between classes between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a difference in directions associated with the pixels in a block area;

calculating a mean of the distance between classes between the block areas for each block of the sample and master data to determine a distance between classes between each block of the sample and master data;

comparing the distance between classes between each block of the sample and master data to a reference distance between classes; and

verifying a fingerprint as that of the designated person based on results from the comparisons.

48. The method of claim 37, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination nito a plurality of block areas;

dividing each of the blocks into a plurality of block areas;

determining a dispersion within a class of each block of the sample image based on the sample data which represents a direction of a finger characteristic within a block;

determining a dispersion within a class of each block of the master image based on the master data which represents a direction of a finger characteristic within a block;

determining the aberration of direction between blocks of the sample image and master image representing a difference in a finger characteristic between respective blocks of the sample image and master image based on the sample data and master data;

determining a total dispersion based on the dispersion within a class for each block of the sample image and the master image and the aberration of direction which represents a measure of the similarity between the sample image and the master image;

determining a cross-correlation between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a similarity in directions associated with the pixels in a block area;

calculating a mean of the cross-correlations between the block areas for each block of the sample and master image to determine a cross-correlation between each block of the sample and master image;

first comparing the cross-correlations between such block of the sample and master image to a reference cross-correlation;

second comparing the total dispersion for each block to a reference total dispersion; and

verifying a fingerprint as that of the designated person based on results from the first and second comparisons.

49. The method of claim 37, wherein the fingerprint verification step comprises the following steps:

dividing the areas of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block areas;

determining a dispersion within a class of each block of the sample image based on the sample data which represents a direction of a finger characteristic within a block;

determining a dispersion within a class of each block of the master image based on the master data which represents a direction of a finger characteristic within a block;

determining the aberration of direction between blocks of the sample image and master image representing a difference in a finger characteristic between respective blocks of the sample image and master image based on the sample data and master data;

determining a total dispersion based on the dispersion within a class for each block of the sample image and the master image and the aberration of direction which represents a measure of the similarity between the sample image and the master image;

determining a distance between classes between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a difference in directions associated with the pixels in a block area;

calculating a mean of the distance between classes between the block areas for each block of the sample and master data to determine a distance between classes between each block of the sample and master data;

first comparing the distance between classes between each block of the sample and master data to a reference distance between classes;

second comparing the total dispersion for each block to a reference total dispersion; and

verifying a fingerprint as that of the designated person based on results from the first and second comparisons.

50. The method of claim 37, wherein the fingerprint verification step comprises the following steps;

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block areas;

determining a cross-correlation between the blocks areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a similarity in directions associated with the pixels in a block area;

calculating a mean of the cross-correlations between the block areas for each block of the sample and master image to determine a cross-correlation between each block of the sample and master image;

determining a distance between classes between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a difference in directions associated with the pixels in a block area;

calculating a mean of the distance between classes between the block areas for each block of the sample and master data to determine a distance between classes between each block of the sample and master data;

first comparing the distance between classes between each block of the sample and master data to a reference distance between classes;

second comparing the cross-correlations between each block of the sample and master image to a reference cross-correlation;

verifying a fingerprint as that of the designated person based on results from the first and second comparisons.

51. The method of claim 37, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block area;

p1 determining a dispersion within a class of each block of the sample image based on the sample data which represents a direction of a finger characteristic within a block;

determining a dispersion within a class of each block of the master image based on the master data which represents a direction of a finger characteristic within a block;

determining the aberration of direction between blocks of the sample image and master image representing a difference in a finger characteristic between respective blocks of the sample image and master image based on the sample data and master data;

determining a total dispersion based on the dispersion within a class for each block of the sample image and the master image and the aberration of direction which represents a measure of the similarity between the sample image and the master image;

determining a cross-correlation between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a similarity in directions associated with the pixels in a block area;

calculating a mean of the cross-correlations between the block areas for each block of the sample and master image to determine a cross-correlation between each block of the sample and master image;

determining a distance between classes between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a difference in directions associated with the pixels in a block area;

calculating a mean of the distance between classes between the block areas for each block of the sample and master data to determine a distance between classes between each block of the sample and master data;

first comparing the distance between classes between each block of the sample and master data to a reference distance between classes;

second comparing the cross-correlations between each block of the sample and master image to a reference cross-correlation;

third comparing the total dispersion for each block to a reference total dispersion; and

verifying a fingerprint as that of the designated person based on results from the first, second and third comparisons.

52. A method for generating sample and master data for fingerprint verification, the method comprising the steps of:

inputting a gray scale image, which is one of the master image or the sample image, of a finger including a fillet center of said finger using an image processing portion, said gray scale image composed of pixels;

outputting the gray scale image from said image processing portion to a data processing portion;

determining a center of characteristics of said gray scale image using the data processing portion;

taking as an area of examination an area having said center of characteristics as its center and including said fillet center using the data processing portion;

calculating density partial differentials for each pixel in the area of examination between said pixel and at least adjacent pixels in a plurality of directions using the data processing portion;

determining the master data or the sample data in the data processing portion, the master data or sample data being a direction of each pixel in the area of examination of the master image or sample image respectively, the direction of a pixel being a minimum density partial differential for each pixel in the area of examination; and

performing fingerprint verification based on the sample and master data in the data processing portion.

53. The method of claim 52, wherein after the step of taking, the method comprises the additional steps of:

dividing the area of examination into a plurality of blocks in the image processing portion; and

determining a direction of a block for the master data or the sample data in the image processing portion, the direction of a block being the direction of a majority of the pixels in the block.

54. The method of claim 52, wherein after the step of taking, the method comprises the additional steps of:

dividing the area of examination into a plurality of blocks in the image processing portion;

dividing each block into a plurality of block areas in the image processing portion; and

determining a direction of a block area for the master data or the sample data in the image processing portion, the direction of a block area being the direction of a majority of the pixels in the block area.

55. The method as in claim 52, wherein the step of determining the center of characteristics comprises the steps of:

determining partial differentials of density values between adjacent pixels in both vertical and horizontal directions of the gray scale image;

p1 making histograms of the determined partial differentials of density values in the vertical and horizontal directions; and

determining the center of characteristics of the gray scale image based on the histograms.

56. The method as in claim 52, wherein the step of determining the center of characteristics comprises the steps of:

taking partial differential of adjacent pixels in the gray scale image in both a vertical, y coordinate axis, direction and horizontal, x coordinate axis, direction,

summing the partial differentials between pixels parallel to the y-axis for each coordinate of the x-axis,

determining as an x-max coordinate, a coordinate on the x-axis corresponding to a sum of partial differentials parallel to the y-axis which is a maximum,

summing the partial differentials between pixels parallel to the x-axis for each coordinate of the y-axis,

determining as a y-max coordinate, a coordinate on the y-axis corresponding to a sum of partial differentials parallel to the x-axis which is a maximum, and

designating as the center of characteristics for the gray scale image the x-max and y-max coordinates.

57. The method of claim 52, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block areas;

determining a dispersion within a class of each block of the sample image based on the sample data which represents a direction of a finger characteristic within a block;

determining a dispersion within a class of each block of the master image based on the master data which represents a direction of a finger characteristic within a block;

determining the aberration of direction between blocks of the sample image and master image representing a difference in a finger characteristics between respective blocks of the sample image and master image based on the sample data and master data;

determining a total dispersion based on the dispersion within a class for each block of the sample image and the master image and the aberration of direction which represents a measure of the similarity between the sample image and the master image;

comparing the total dispersion for each block of a reference total dispersion; and

verifying a fingerprint as that of the designated persons based on results from the comparisons.

58. The method as in claim 57, wherein the steps of determining the dispersions within a class for each block of the sample image and master image comprises the steps of:

determining an aberration of direction between a plurality of direction codes and a direction of a pixel for each pixel in each block area which represents a difference in direction between a direction of a pixel and a direction code;

determining a dispersion within a class of each block area for each direction code, for each direction and code the determination of a dispersion within a class of each block area comprises the steps of,

multiplying each pixel within a block area by a weight to form a pixel product, the weight corresponding to the aberration of direction between the pixel and the direction code;

summing the pixel products of the pixels in the block area;

dividing the summed pixel products for the block area by the number of pixels in the block area to determine a dispersion within a class of the block area for that direction code;

determining as the dispersion within a class of a block area a one of the dispersion within a class of the block area for each direction code which is a minimum;

summing as a block dispersion sum, the dispersion within a class of a block area for a block; and

dividing the block sum of a block by a number of block areas in the block to determine the dispersion within a class of the block.

59. The method as in claim 58, wherein the step of determining the aberration of direction comprises the steps of:

providing for each block area of the sample and master image a direction associated with a fingerprint characteristic of a block area, the provided direction being the direction code for which the dispersion within a class of the block area for the direction codes was a minimum;

determining a direction, which is associated with a fingerprint characteristic of a block, for each block in the sample and master image by calculating the mean of the provided directions of each block area within a respective block;

determining an aberration of direction between corresponding blocks of the sample and master image based on the determined directions of the blocks in the sample and master image.

60. The method as in claim 57, wherein the step of determining the aberration of direction comprises the steps of:

determining a direction, which is associated with a fingerprint characteristic of a block area, for each block area in a block for both the sample and master image based on the sample and master data;

determining a direction, which is associated with a fingerprint characteristic of a block, for each block in the sample and master image by calculating the mean of the directions of each block area within a respective block;

determining an aberration of direction between corresponding blocks of the sample and master image based on the determined directions of the blocks in the sample and master image.

61. The method of claim 52, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block areas;

determining a cross-correlation between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a similarity in directions associated with the pixels in a block area;

calculating a mean of the cross-correlations between the block areas for each block of the sample and master image to determine a cross-correlation between each block of the sample and master image;

comparing the cross-correlations between each block of the sample and master image to a reference cross-correlation; and

verifying a fingerprint as that of the designated person based on result from the comparisons.

62. The method of claim 52, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block areas;

determining a distance between classes between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a difference in directions associated with the pixels in a block area;

calculating a mean of the distance between classes between the block areas for each block of the sample and master data to determine a distance between classes between each block of the sample and master data;

comparing the distance between classes between each block of the sample and master data to a reference distance between classes; and

verifying a fingerprint as that of the designated person based on results from the comparisons.

63. The method of claim 52, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block areas;

determining a dispersion within a class of each block of the sample image based on the sample data which represents a direction of a finger characteristic within a block;

determining a dispersion within a class of each block of the master image based on the master data which represents a direction of a finger characteristic within a block;

determining the aberration of direction between blocks of the sample image and master image representing a difference to a finger characteristic between respective blocks of the sample image and master image based on the sample data and master data;

determining a total dispersion based on the dispersion within a class for each block of the sample image and the master image and the aberration of direction which represents a measure of the similarity between the sample image and the master image;

determining a cross-correction between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a similarity in directions associated with the pixels in a block area;

calculating a mean of the cross-correlation between the block areas for each block of the sample and master image to determine a cross-correlation between each block of the sample and master image;

first comparing the cross-correlations between each block of the sample and master image to a reference cross-correlation;

second comparing the total dispersion for each block to a reference total dispersion; and

verifying a fingerprint as that of the designated person based on results from the first and second comparisons.

64. The method of claim 52, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination into a plurality of blocks;

`dividing each of the blocks into a plurality of block areas;

determining a dispersion within a class of each block of the sample image based on the sample data which represents a direction of a finger characteristic within a block;

determining a dispersion within a class of each block of the master image based on the master data which represents a direction of a finger characteristic within a block;

determining the aberration of direction between blocks of the sample image and master image representing a difference in a finger characteristic between respective blocks of the sample image and master image based on the sample data and master data;

determining a total dispersion based on the dispersion within a class for each block of the sample image and the master image and the aberration of direction which represents a measure of the similarity between the sample image and the master image;

determining a distance between classes between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a difference in directions associated with the pixels in a block area;

calculating a mean of the distance between classes between the block areas for each block of the sample and master data to determine a distance between classes between each block of the sample and master data;

first comparing the distance between classes between each block of the sample and master data to a reference distance between classes;

second comparing the total dispersion for each block to a reference total dispersion; and

verifying a fingerprint as that of the designated person based on results from the first and second comparisons.

65. The method of claim 52, wherein the fingerprint verification step comprises the following steps:

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block areas;

determining a cross-correlation between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a similarity in directions associated with the pixels i a block area;

calculating a mean of the cross-correlations between the block areas for each block of the sample and master image to determine a cross-correlation between each block of the sample and master image;

determining a distance between classes between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a difference in directions associated with the pixels in a block area;

calculating a mean of the distance between classes between the block areas for each block of the sample and master data to determine a distance between classes between each block of the sample and master data;

first comparing the distance between classes between each block of the sample and master data to a reference distance between classes;

second comparing the cross-correlation between each lock of the sample and master image to a reference cross-correlation;

verifying a fingerprint as that of the designated person based on results from the first and second comparisons.

66. The method of claim 52, wherein the fingerprint verification step comprises the followings steps:

dividing the area of examination into a plurality of blocks;

dividing each of the blocks into a plurality of block areas;

determining a dispersion within a class of each block of the sample image based on the sample data which represents a direction of a finger characteristic within a block;

determining a dispersion within a class of each block of the master image based on the master data which represents a direction of a finger characteristic within a block;

determining the aberration of direction between blocks of the sample image and master image representing a difference in a finger characteristic between respective blocks of the sample image and master based on the sample data and master data;

determining a total dispersion based on the dispersion within a class for each block of the sample image and the master image and the aberration of direction which represents a measure of the similarity between the sample image and the master image;

determining a cross-correlation between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a similarity in directions associated with the pixels in a block area;

calculating a mean of the cross-correlations between the block areas for each block of the sample and master image to determine a cross-correlation between each block of the sample and master image;

determining a distance between classes between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a difference in directions associated with the pixels in a block area;

calculating a mean of the distance between classes between the block areas for each block of the sample and master data to determine a distance between classes between each block of the sample and master data;

first comparing the distance between classes between each block of the sample and master data to a reference distance between classes;

second comparing the cross-correlations between each block of the sample and master image to a reference cross-correlation;

third comparing the total dispersion for each block to a reference total dispersion; and

verifying a fingerprint as that of the designated person based on results from the first, second and third comparisons.

67. A method for verifying that a sample image of a fingerprint is that of a designated person through comparison of sample data, generated from the sample image, with master data, generated from a master image of the designated person'"'"'s fingerprint, the master and sample image having been divided into a plurality of blocks and each block having been divided into a plurality of block areas, each block area is comprised of pixels each having an associated direction as the sample and master data, the method comprising the steps of:

determining a dispersion within a class of each block of the sample image based on the sample data which represents a direction of a finger characteristic within a block;

determining a dispersion within a class of each block of the master image based on the master data which represents a direction of a finger characteristic within a block;

determining the aberration of direction between blocks of the sample image and master image representing a difference in a finger characteristic between respective blocks of the sample image and master image based on the sample data and master data;

determining a total dispersion based on the dispersion within a class for each block of the sample image and the master image and the aberration of direction which represents a measure of the similarity between the sample image and the master image;

determining a cross-correlation between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a similarity in directions associated with the pixels in a block area;

calculating a mean of the cross-correlation between the block areas for each block of the sample and master image to determine a cross-correlation between each block of the sample and master image;

determining a distance between classes between the block areas of each block in the sample image and the master image based on a number of pixels in each block area of the sample image and master image associated with a direction which represents a difference in directions associated with the pixels in a block area;

calculating a mean of the distance between classes between the block areas for each block of the sample and master data to determine a distance between classes between each block of the sample and master data;

first comparing the distance between classes between each block of the sample and master data to a reference distance between classes;

second comparing the cross-correlation between each block of the sample and master image to a reference cross-correlation;

third comparing the total dispersion for each block to a reference total dispersion; and

verifying a fingerprint as that of the designated person based on results from the first, second and third comparisons.