Granted Patent B2

US 12,646,349 · App. 18/785,921 · Granted Jun 2, 2026

Inventors: Feng-Chien Hsieh (Pingtung County, TW); Yun-Wei Cheng (Taipei City, TW); Kuo-Cheng Lee (Tainan City, TW); Cheng-Ming Wu (Tainan City, TW); Wei-Li Hu (Tainan City, TW)

Assignee: TAIWAN SEMINCONDUCTOR MANUFACTURING COMPANY, LTD.

G06V40/1318G06V40/1353G06V40/1365H10F39/198H10F39/8053H10F39/8057H10F39/8063

Abstract

Claims (50)

1 . A method of fingerprint verification, comprising:

capturing a fingerprint image by an image sensing device, the image sensing device including a pixel array of a combination of sensing pixels configured to capture minutia points in the fingerprint image and positioning pixels configured to provide positioning codes, wherein the positioning pixels are dark pixels;

calculating vectors of the minutia points with reference to the positioning codes;

comparing the vectors to reference vectors generated from a reference fingerprint image, wherein the positioning codes are insensitive to information in either the fingerprint image or the reference fingerprint image; and

determining a match between the fingerprint image and the reference fingerprint image based on the comparing of the vectors.

2 . The method of claim 1 , wherein the determining of the match includes:

determining shifts between the vectors and the reference vectors; and

determining whether the shifts are substantially equal to conclude the match.

3 . The method of claim 1 , wherein the vectors include a first group of vectors with reference to a first type of the positioning codes and a second group of vectors with reference to a second type of the positioning codes.

4 . The method of claim 3 , wherein the determining of the match includes:

determining first shifts between the first group of vectors with corresponding ones of the reference vectors;

determining whether the first shifts are substantially equal to conclude a first low level match;

determining second shifts between the second group of vectors with corresponding ones of the reference vectors; and

determining whether the second shifts are substantially equal to conclude a second low level match,

wherein the match is concluded upon at least conclusion of both the first low level match and the second low level match.

5 . The method of claim 4 , wherein the determining of the match further includes:

determining whether the first shifts and the second shifts are substantially equal to conclude a high level match,

wherein the match is concluded additionally upon conclusion of the high level match.

6 . The method of claim 1 , further comprising:

comparing the minutia points in the fingerprint image to minutia points in the reference fingerprint image,

wherein the determining of the match is additionally based on the comparing of the minutia points.

7 . The method of claim 1 , wherein the determining of the match is solely based on the comparing of the vectors.

8 . The method of claim 1 , wherein the sensing pixels include color pixels configured to provide a skin tone of the fingerprint image, the method further comprising:

comparing the skin tone of the fingerprint image to a reference skin tone of the reference fingerprint image.

9 . The method of claim 8 , wherein the determining of the match is additionally based on the comparing of the skin tone.

10 . A method of biometric sensing, comprising:

acquiring an image of an object by a biometric sensing device, the biometric sensing device including a pixel array of a combination of sensing pixels and positioning pixels, wherein the positioning pixels are dark pixels;

retrieving minutia points from the acquired image;

calculating vectors of the minutia points with reference to the positioning pixels;

calculating shifts by comparing the vectors to reference vectors generated from a reference image, wherein the positioning pixels are insensitive to information in either the acquired image or the reference image; and

determining a match between the acquired image and the reference image by evaluating consistency of the shifts.

11 . The method of claim 10 , wherein the determining of the match is based on the shifts being substantially equal.

12 . The method of claim 10 , wherein each of the shifts is in format of a vector.

13 . The method of claim 10 , wherein the positioning pixels include a first subset corresponding to a first type of pattern arrangement and a second subset corresponding to a second type of pattern arrangement.

14 . The method of claim 13 , wherein the calculating of the shifts includes:

calculating a first group of the shifts corresponding to the first subset of the positioning pixels; and

calculating a second group of the shifts corresponding to the second subset of the positioning pixels.

15 . The method of claim 14 , wherein the evaluating consistency of the shifts includes:

evaluating a first consistency of the first group of the shifts; and

evaluating a second consistency of the second group of the shifts.

16 . A method of fingerprint verification, comprising:

capturing a fingerprint image by an image sensing device, the image sensing device including a pixel array of a combination of monochromatic pixels configured to capture a minutia map of the fingerprint image, color pixels configured to capture a skin tone of the fingerprint image, and positioning pixels configured to provide vectors of minutia points in the minutia map with reference to the positioning pixels;

comparing the minutia map to a reference minutia map of a reference fingerprint image;

comparing the skin tone to a reference skin tone of the reference fingerprint image;

evaluating consistency of shifts between the vectors and reference vectors of the reference fingerprint image; and

determining a match between the fingerprint image and the reference fingerprint image based on the comparing of the minutia map, the comparing of the skin tone, and the evaluating of the consistency.

17 . The method of claim 16 , wherein the monochromatic pixels outnumbers either of the color pixels and the positioning pixels.

18 . The method of claim 16 , wherein the color pixels outnumber the monochromatic pixels, and the monochromatic pixels outnumber the positioning pixels.

19 . The method of claim 16 , wherein the color pixels are positioned adjacent to the positioning pixels.

20 . The method of claim 19 , wherein the color pixels are arranged in a pattern same to a pattern formed by the adjacent positioning pixels.

Assignments (1)

Continuity (4)

References Cited (34)