Fingerprint identification system

• US 5,926,555 A
• Filed: 04/03/1997
• Issued: 07/20/1999
• Est. Priority Date: 10/20/1994
• Status: Expired due to Fees

• Current Assignee: Veridian Engineering Incorporated
• Original Assignee: Calpine Corporation
• Inventors: Kiefer, Frederick W., Dennison, Raymond J., Ort, James R., Lange, Douglas L.
• Primary Examiner(s): Tran, Phuoc
• Application Number:
  US08/834,547
• Time in Patent Office: 838 Days
• Field of Search: 382/124, 382/125, 382/203, 382/204, 382/260, 382/266, 382/286, 382/257
• US Class Current: 382/124
• CPC Class Codes: G06V 40/1347 Preprocessing; Feature extraction
• CPC Class Codes: G06V 40/1359 Extracting features related to ridge properties; Determining the fingerprint type, e.g. whorl or loop
• CPC Class Codes: G07C 9/37 using biometric data, e.g. fingerprints, iris scans or voice recognition

Abstract

A fingerprint identification system from a gray scale image of the fingerprint includes the steps of converting fingerprint image into a digitized gray scale image, enhancing gray scale to provide uniform contrast over entire fingerprint image, down-sampling the contrast enhanced image to limit the number of pixels analyzed, generating a gradient map of the fingerprint image, defining a calculation kernel over which ridge angle determination is made by reference to gradient map, causing the calculation kernel to traverse over gradient map in a predetermined increment over pixels comprising the fingerprint image, smoothing the ridge angle map using the process of Fourier filtering, and smoothing the image quality map using the process of erosion and dilation. The ridge angle map and image quality maps are recorded. The frequency of ridges of a fingerprint image is determined by establishing at each selected pixel a line perpendicular to ridge flow, establishing a plurality of lines parallel to perpendicular line placed on both sides of perpendicular line of selected pixel, sampling a predetermined number of pixels along the perpendicular lines, averaging sample of each line, determining frequency of ridges by calculating power spectrum from a Fourier transform.

Claims

1. Apparatus for generating readable record of a fingerprint image including

means for converting a fingerprint image into a digitized gray scale image comprising a plurality of individual pixels;

means for determining the ridge angle and image quality of the fingerprint image at a plurality of regularly spaced individual pixels to create a Ridge Angle Map and an image quality map; and

means for recording said Ridge Angle Map at those locations where the image quality is above a predetermined level, and

means for recording said image quality map of said fingerprint image.

2. The apparatus of claim 1 wherein said image converting means includes

means for enhancing the gray scale to provide a uniform contrast over the entire fingerprint image.

3. The apparatus of claim 2 further including

means for down-sampling the contrast enhanced image to limit the number of pixels analyzed.

4. The apparatus of claim 1 wherein said means for determining the ridge angle and said means for determining image quality of the fingerprint image includes

means for generating a gradient map of the fingerprint image, and

means for defining a calculation kernel over which the ridge angle determination is made by reference to said gradient map;

and

means for causing the calculation kernel to traverse over said gradient map in a predetermined increment over the pixels comprising the fingerprint image.

5. The apparatus of claim 4 wherein said predetermined increments are an integer number greater than or equal to one.

6. The apparatus of claim 5 including means for assigning the same ridge angle and quality values assigned to each of said plurality of pixels analyzed to surrounding pixels.

7. The apparatus of claim 1 further including

means for determining the frequency of the ridges in a fingerprint at said regularly spaced intervals.

8. A method for generating a machine readable record of an image of a fingerprint, said method comprising the steps of

converting a fingerprint image into a digitized gray scale image comprising a plurality of individual pixels;

determining the ridge angle and image quality of the fingerprint image at regularly spaced individual pixels to create a Ridge Angle Map and an image quality map; and

recording said Ridge Angle Map and said image quality map of said fingerprint image.

9. The method of claim 8 further comprising the steps of

means for enhancing the gray scale to provide a uniform contrast over the entire fingerprint image;

and

down-sampling the contrast enhanced image to limit the number of pixels analyzed.

10. The method of claim 8 wherein the step of determining the ridge angle and quality of the fingerprint image includes the steps of

generating a gradient map of the fingerprint image, and

defining a calculation kernal over which the ridge angle determination is made with reference to said gradient map;

and

traversing the calculation kernal over said gradient map in predetermined increments over the pixels comprising the fingerprint image.

11. The method of claim 8 wherein the step of determining the ridge angle and image quality of the fingerprint image at regularly spaced individual pixels to create a Ridge Angle Map and an image quality map is performed by the Direct Estimation of Static Fields process.

12. The method of claim 8 further including the step of smoothing the Ridge Angle Map and the image quality map.

13. The method of claim 12 wherein the step of smoothing the Ridge Angle Map is performed by the process of erosion and dilation.

14. The method of claim 12 wherein the step of smoothing the image quality map is performed by the process of Fourier filtering.

15. The method of claim 8 further including the step of determining the frequency of the ridges in a fingerprint at said regularly spaced intervals to create a Ridge Frequency Map.

16. The method of claim 15 further including the step of smoothing the Ridge Frequency Map.

17. The method of claim 16 wherein the smoothing step is performed by Fourier filtering.

18. A method of generating a machine readable record of an image of a fingerprint comprising the steps of

converting a fingerprint image into a digitized gray scale image comprising a plurality of individual pixels;

enhancing the gray scale to provide a uniform contrast over the entire fingerprint image; and

down-sampling the contrast enhanced image to limit the number of pixels to be analyzed;

determining the ridge angle and image quality of the fingerprint image at regularly spaced individual pixels to create a Ridge Angle Map and an image quality map, including the steps of

generating a gradient map of the fingerprint image, and

defining a calculation kernal over which the ridge angle determination

is made with reference to said gradient map; and

traversing the calculation kernal over said gradient map in predetermined increments over the pixels comprising the fingerprint image; and

recording said Ridge Angle Map and said image quality map of said fingerprint image.

19. The method of claim 18 further including the step of smoothing the Ridge Angle Map and the image quality map.

20. The method of claim 19 wherein the step of smoothing the Ridge Angle Map is performed by the process of erosion and dilation.

21. The method of claim 19 wherein the step of smoothing the image quality map is performed by the process of Fourier filtering.

22. The method of claim 18 further including the step of determining the frequency of the ridges in a fingerprint at said regularly spaced intervals to create a Ridge Frequency Map.

23. The method of claim 22 wherein the step of determining the frequency of the ridges includes the steps of

establish at each selected pixel a line perpendicular to the ridge flow,

sample a predetermined number of pixels on both sides of said perpendicular line, and

determine the frequency of the ridges by calculating a power spectrum from a Fourier transform.

24. The method of claim 23 further comprising the steps of

establishing a plurality of lines parallel to said perpendicular lines and placed on both sides of said perpendicular line selected pixel,

sampling a predetermined number of equally spaced pixels along both said perpendicular line and said plurality of lines, and

averaging the samples of each line prior to determining the frequency.

25. The method of claim 18 further including the step of smoothing the Ridge Frequency Map.

26. The method of claim 25 wherein the smoothing step is performed by Fourier filtering.