Fingerprint analyzing and encoding system

• US 6,002,787 A
• Filed: 12/06/1995
• Issued: 12/14/1999
• Est. Priority Date: 10/27/1992
• Status: Expired due to Term

• Current Assignee: 123 ID Incorporated
• Original Assignee: Jasper Consulting, Inc. (Minnesota)
• Inventors: Wendt, Barry M., Wittig, Benedict A., Takhar, Harinder S.
• Primary Examiner(s): CHANG, JON CARLTON
• Application Number:
  US08/568,404
• Time in Patent Office: 1,469 Days
• Field of Search: 382/124, 382/125, 382/126, 382/127, 283/68, 283/69, 356/71, 396/15
• US Class Current: 382/125
• CPC Class Codes: G06Q 20/4037 Remote solvency checks
• CPC Class Codes: G06V 40/1347 Preprocessing; Feature extraction
• CPC Class Codes: G07C 9/257 electronically G07C9/26 takes precedence
• CPC Class Codes: G07F 7/08 by coded identity card or credit card or other personal identification means without personal verification means G07F7/02
• CPC Class Codes: G07F 7/10 together with a coded signal , e.g. in the form of personal identification information, like personal identification number [PIN] or biometric data

Abstract

A system for converting an image-enhanced digitized raster fingerprint image to vector lines in order to generate a unique identification value for the fingerprint. The raster image pixels are converted to vector lines along the fingerprint ridges and the vector lines are classified and converted according to type. The line types are then analyzed and a list of identification features corresponding to the vector line types is generated. The identification features between the vector line types are compared and the image is classified according to fingerprint class. A unique identification value is then generated by numerically encoding the classified identification features.

Claims

1. A method of converting an image-enhanced digitized raster fingerprint image having raster image pixels to vector line types in order to generate a unique identification number for the fingerprint, comprising the steps of:

(a) electronically converting the raster image pixels to a plurality of vectorized lines, each vectorized ridge line comprising a plurality of vector line segments along the fingerprint ridges;

(b) electronically classifying each of the vectorized ridge lines as belonging to one of a plurality of vectorized ridge line types;

(c) analyzing the vectorized ridge lines and generating a list of identification features corresponding to each of the particular vectorized ridge lines;

(d) classifying the image as belonging to one of a plurality of fingerprint classes by comparing the identification features corresponding to the vectorized ridge lines; and

(e) using the classified image to generate a fingerprint identification code by numerically encoding the identification features and line types corresponding to the vectorized ridge lines of the image, wherein the plurality of vectorized ridge line types includes line, loop, double loop, wave, spiral, oval, partial arc, and bifurcation types.

2. The method of claim 1 wherein the plurality of fingerprint classes includes plain arch, tented arch, radial loop, ulnar loop, accidental whorl, central pocket loop whorl, double loop whorl, and plain whorl.

3. A method of converting an image-enhanced digitized raster fingerprint image to vector line types in order to generate a unique identification number for the fingerprint, comprising the steps of:

(a) electronically defining vectorized ridge lines, each vectorized ridge line comprising a plurality of vector line segments along the fingerprint ridges, the vectorized ridge lines corresponding to a plurality of vectorized ridge line types within the fingerprint;

(b) classifying the fingerprint by class according to the line types;

(c) electronically locating the coordinates of a core point, delta point, and origin point of the fingerprint;

(d) electronically determining a geometric mean line of the fingerprint according to the vectorized ridge lines of the fingerprint classification and calibrating the vectorized ridge lines to a fixed coordinate system in order to reassign an origin;

(e) determining identification features of the fingerprint from the core point, delta point, and origin point; and

(f) generating a fingerprint identification code by numerically encoding the identification features and line types of the fingerprint.

4. The method of claim 3 wherein the plurality of vectorized ridge line types includes line, loop, double loop, wave, spiral, oval, partial arc, and bifurcation types.

5. The method of claim 3 wherein the plurality of fingerprint classes includes plain arch, tented arch, radial loop, ulnar loop, accidental whorl, central pocket loop whorl, double loop whorl, and plain whorl.

6. The method of claim 3 further comprising the steps of:

(a) projecting a first vector from the core point parallel to the geometric mean line of the fingerprint image;

(b) projecting a second vector from the delta point to a point intersecting with the first vector;

(c) defining a third vector connecting the core point and the delta point; and

(d) determining a first ridge count along the first vector, a second ridge count along the second vector, and a third ridge count long the third vector;

wherein the first, second, and third ridge counts comprise identification features of the image used to generate the fingerprint identification code.

7. The method of claim 6 wherein the identification features of the fingerprint include the number of ridge crossings of the first, second, and third vectors, the length and angle of the core coordinate vector, the delta coordinate, and the fingerprint classification.

8. The method of claim 6 wherein the triangle comprises a right triangle wherein the second vector intersects with the first vector at a substantially right angle.

9. The method of claim 3 wherein the step of calibrating the vectorized ridge lines comprises the step of normalizing features of the fingerprint image to a fixed coordinate system having a x-axis and a y-axis.

10. The method of claim 9 further comprising the steps of:

(a) determining a geometric mean for each of the plurality of vectorized ridge lines within the fingerprint image and averaging the geometric mean for each of the plurality of vectorized ridge lines to determine a geometric mean line of the fingerprint image;

(b) projecting a first vector from the core point to a point parallel to the geometric mean line of the fingerprint image;

(c) projecting a second vector from the delta point to a point intersecting with the first vector, the intersection point forming an origin;

(d) defining a third vector connecting the core point and the delta point;

(e) rotating the fingerprint image around the origin point until the second vector is parallel with the x-axis and the first vector is parallel with the y-axis; and

(f) determining a first ridge count along the first vector, a second ridge count along the second vector, and a third ridge count long the third vector;

wherein the first, second, and third ridge counts comprise identification features of the image used to generate the fingerprint identification code.

11. The method of claim 9 wherein the features of the fingerprint image include minutiae points, bifurcation points, and complex vectors.

12. A method for verifying that a sample image of a fingerprint is that of a designated person, the sample image being an image-enhanced digitized raster image having raster image pixels, the method comprising the steps of:

(a) electronically converting the raster image pixels to a plurality of vectorized ridge lines, each vectorized ridge line comprising a plurality of vector line segments along the fingerprint ridges;

(b) electronically classifying each of the vectorized ridge lines as belonging to one of a plurality of vector line types;

(c) electronically identifying each of the vector lines corresponding to its classified line type and identifying each of the vector lines as belonging to a particular vectorized ridge line;

(d) analyzing the vectorized ridge lines and generating a list of identification features corresponding to the vectorized ridge lines;

(e) classifying the image as belonging to one of a plurality of fingerprint classes by comparing the identification features corresponding to the vector line types;

(f) using the classified image to generate a fingerprint identification code by numerically encoding the identification features and line types corresponding to the vectorized ridge lines of the image; and

(g) comparing the fingerprint identification code of the sample image with that of the fingerprint identification code of the designated person, wherein the plurality of vector line types includes line, loop, double loop, wave, spiral, oval, partial arc, and bifurcation types.

13. The method of claim 12 wherein the plurality of fingerprint classes includes plain arch, tented arch, radial loop, ulnar loop, accidental whorl, central pocket loop whorl, double loop whorl, and plain whorl.

14. A method of classifying an image-enhanced rasterized fingerprint image according to fingerprint class comprising the steps of:

(a) electronically converting pixels of the rasterized fingerprint image to a plurality of vectorized ridge lines, each vectorized ridge line comprising a plurality of vector line segments along the fingerprint ridges;

(b) electronically classifying each of the vectorized ridge lines as belonging to one of a plurality of vectorized ridge line types;

(c) analyzing the vectorized ridge lines and generating a list of corresponding identification features; and

(d) classifying the image as belonging to one of a plurality of fingerprint classes by comparing the identification features and line types corresponding to the vectorized ridge lines, wherein the plurality of vectorized ridge line types includes line, loop, double loop, wave, spiral, oval, partial arc, and bifurcation types.

15. The method of claim 14 further comprising the step of using the classified image to generate a fingerprint identification code by numerically encoding the identification features and line types of the image.

16. The method of claim 15 further comprising the step of comparing the fingerprint identification code of the sample image with that of the fingerprint identification code of the designated person.

17. The method of claim 14 wherein the plurality of fingerprint classes includes plain arch, tented arch, radial loop, ulnar loop, accidental whorl, central pocket loop whorl, double loop whorl, and plain whorl.

18. The method of claim 14 wherein the identification features of the fingerprint include the number of ridge crossings of a first, second, and third vector, a length and angle of a core coordinate vector, a delta point coordinate, and the fingerprint classification.

19. A method of normalizing the features of an image-enhanced digitized raster fingerprint image to a fixed coordinate system having a x-axis and a y-axis, the method comprising the steps of:

(a) electronically locating the coordinates of a core point and a delta point of the fingerprint;

(b) projecting a first vector from the core point parallel to the geometric mean line of the fingerprint image;

(c) projecting a second vector from the delta point to a point intersecting with the first vector, thereby forming an origin; and

(d) rotating the image around the origin until the second vector is parallel with the x-axis and the first vector is parallel with the y-axis.

20. The method of claim 19 wherein the features of the fingerprint image include minutiae points, bifurcation points, and complex vectors.

21. A method of converting an image-enhanced digitized raster fingerprint image having raster image pixels to vector line types in order to generate a unique identification number for the fingerprint, comprising the steps of:

(a) electronically converting the raster image pixels to a plurality of vectorized lines, each vectorized ridge line comprising a plurality of vector line segments along the fingerprint ridges;

(b) electronically classifying each of the vectorized ridge lines as belonging to one of a plurality of vectorized ridge line types;

(c) analyzing the vectorized ridge lines and generating a list of identification features corresponding to each of the particular vectorized ridge lines;

(d) classifying the image as belonging to one of a plurality of fingerprint classes by comparing the identification features corresponding to the vectorized ridge lines; and

(e) using the classified image to generate a fingerprint identification code by numerically encoding the identification features and line types corresponding to the vectorized ridge lines of the image;

(f) determining a geometric mean for each of the plurality of vectorized ridge lines within the fingerprint image and averaging the geometric mean for each of the plurality of vectorized ridge lines to determine a geometric mean line of the fingerprint image;

(g) projecting a first vector from a core point parallel to the geometric mean line of the fingerprint image;

(h) projecting a second vector from a delta point to a point intersecting with the first vector;

(i) defining a third vector connecting the core point and the delta point; and

(j) determining a first ridge count along the first vector, a second ridge count along the second vector, and a third ridge count long the third vector;

wherein the first, second, and third ridge counts comprise identification features of the image used to generate the fingerprint identification code.

22. The method of claim 21 wherein the identification features of the fingerprint include the number of ridge crossings of the first, second, and third vectors, the length and angle of a core coordinator vector, the delta point coordinate, and the fingerprint classification.

23. The method of claim 21 wherein the second vector intersects with the first vector at a substantially right angle.

24. The method of claim 21 wherein the step of numerically encoding the identification features of the fingerprint image further comprises the step of concatenating at least two features selected from the group consisting of a delta point coordinate value, core point coordinate value, ridge count values for the first, second, and third vectors, and an image classification value into the unique identification number.

25. The method of claim 24 wherein the delta point coordinate value and the core point coordinate value each comprise an 11-bit signed number ranging from -999 to +999.

26. The method of claim 24 wherein ridge count values for the first, second, and third vectors each comprise a 6-bit integer number ranging from 0 to 63.

27. The method of claim 24 wherein the image classification value comprises a 2-bit integer number where a value of 0 represents a loop, 1 represents a whorl, and 2 represents an arch.

28. The method of claim 24 wherein the step of numerically encoding the identification features of the fingerprint image further comprises the step of concatenating a list of a plurality of vector lines for at least a portion of the vectorized ridge lines with the unique identification value.

29. A method for verifying that a sample image of a fingerprint is that of a designated person, the sample image being an image-enhanced digitized raster image having raster image pixels, the method comprising the steps of:

(a) electronically converting the raster image pixels to a plurality of vectorized ridge lines, each vectorized ridge line comprising a plurality of vector line segments along the fingerprint ridges;

(b) electronically classifying each of the vectorized ridge lines as belonging to one of a plurality of vector line types;

(c) electronically identifying each of the vector lines corresponding to its classified line type and identifying each of the vector lines as belonging to a particular vectorized ridge line;

(d) analyzing the vectorized ridge lines and generating a list of identification features corresponding to the vectorized ridge lines;

(e) classifying the image as belonging to one of a plurality of fingerprint classes by comparing the identification features corresponding to the vectorized ridge lines;

(f) using the classified image to generate a fingerprint identification code by numerically encoding the identification features and line types corresponding to the vectorized ridge lines of the image;

(g) comparing the fingerprint identification code of the sample image with that of the fingerprint identification code of the designated person;

(i) determining a geometric mean for each of the plurality of vectorized ridge lines within the fingerprint image and averaging the geometric mean for each of the plurality of vectorized ridge lines to determine a geometric mean line of the fingerprint image;

(j) projecting a first vector from a core point to a point parallel to the geometric mean line of the fingerprint image;

(k) projecting a second vector from a delta point to a point intersecting with the first vector at a substantially right angle;

(l) defining a third vector connecting the core point and the delta point; and

(m) determining a first ridge count along the first vector, a second ridge count along the second vector, and a third ridge count long the third vector;

wherein the first, second, and third ridge counts comprise identification features of the image used to generate the fingerprint identification code.

30. The method of claim 29 wherein the plurality of vectorized ridge line types includes line, loop, double loop, wave, spiral, oval, partial arc, and bifurcation types.

31. The method of claim 30 wherein the plurality of fingerprint classes includes plain arch, tented arch, radial loop, ulnar loop, accidental whorl, central pocket loop whorl, double loop whorl, and plain whorl.

32. The method of claim 29 wherein the identification features of the fingerprint include the number of ridge crossings of the first, second, and third vectors, the length and angle of a core coordinate vector, the delta point coordinate, and the fingerprint classification.

33. The method of claim 29 wherein the triangle comprises a substantially right triangle.