Method and apparatus for rolled fingerprint capture

• US 7,095,880 B2
• Filed: 09/20/2002
• Issued: 08/22/2006
• Est. Priority Date: 08/19/1999
• Status: Expired due to Term

• Current Assignee: HID Global Corporation (Assa Abloy AB)
• Original Assignee: Cross Match Technologies Incorporated (Assa Abloy AB)
• Inventors: Martinez, Chris J, Scott, Walter Guy
• Primary Examiner(s): AHMED, SAMIR ANWAR
• Application Number:
  US10/247,285
• Publication Number:
  US 20030091219A1
• Time in Patent Office: 1,432 Days
• Field of Search: 382/115, 382/124, 382/125, 382/127, 235/380, 235/492, 340/5.82, 340/5.83, 283/68, 356/71, 396/14, 396/15, 902/3, 902/4, 902/25
• US Class Current: 382/124
• CPC Class Codes: G06V 40/1335 Combining adjacent partial images (e.g. slices) to create a composite input or reference pattern; Tracking a sweeping finger movement

Abstract

A method and apparatus for rolled fingerprint capture is described. The start of a fingerprint roll is detected. A plurality of fingerprint image frames are captured. A centroid window corresponding to each of the plurality of captured fingerprint image frames is determined. Pixels of each determined centroid window are knitted into a composite fingerprint image. The end of the fingerprint roll is detected.

Claims

1. A method for rolled fingerprint capture, comprising the steps of:

(1) detecting the start of a fingerprint roll;

(2) capturing a plurality of fingerprint image frames;

(3) determining a centroid window corresponding to each of the plurality of captured fingerprint image frames, including the steps of;

(a) generating a pixel window in a captured fingerprint image frame;

(b) finding a leading edge column and a trailing edge column of a fingerprint image in the corresponding generated pixel window, including the steps of;

(i) building a histogram representative of the cumulative pixel intensity of pixels present in each column of the generated window;

(ii) scanning through the histogram in the direction opposite that in which the finger is rolled for a difference in the total pixel intensities of first adjacent columns of the generated pixel window that is greater than a first fingerprint edge threshold to find a leading edge; and

(iii) scanning through the histogram in the direction in which the finger is rolled for a difference in the total pixel intensities of second adjacent columns of the generated pixel window that is greater than a second fingerprint edge threshold to find a trailing edge; and

(c) generating the centroid window in the captured fingerprint image frame, wherein the centroid window is bounded by the leading edge column found and the trailing edge column found;

(4) knitting pixels of each determined centroid window into a composite fingerprint image; and

(5) detecting the end of said fingerprint roll.

2. The method of claim 1, further comprising the steps of:

(iv) selecting the column of darker total pixel intensity of the first adjacent columns as the leading edge column; and

(v) selecting the column of darker total pixel intensity of the second adjacent columns as the trailing edge column.

3. An apparatus for rolled fingerprint capture, comprising:

a fingerprint scanner;

a computer system coupled to said fingerprint scanner, wherein said computer system comprises a rolled fingerprint capture module; and

a display that displays a composite fingerprint image generated by said rolled fingerprint capture module, wherein said rolled fingerprint capture module determines a centroid window corresponding to each of a plurality of fingerprint image frames captured by said fingerprint scanner and knits pixels of each said determined centroid window into a composite fingerprint image;

wherein in each of said plurality of fingerprint image frames, said rolled fingerprint capture module generates a pixel window;

wherein said rolled fingerprint capture module comprises a histogram builder;

wherein for each said generated pixel window, said histogram builder builds a histogram representative of the cumulative pixel intensity of pixels present in each column of said generated pixel window;

wherein for each said histogram, said rolled fingerprint capture module scans through said histogram in the direction opposite that in which the finger is rolled for a difference in the total pixel intensities of first adjacent columns of the generated pixel window that is greater than a first fingerprint edge threshold to find a leading edge column;

wherein for each said histogram, said rolled fingerprint capture module scans through said histogram in the direction in which the finger is rolled for a difference in the total pixel intensities of second adjacent columns of the generated pixel window that is greater than a second fingerprint edge threshold to find a trailing edge column; and

wherein said rolled fingerprint capture module generates said determined centroid window in said each of said plurality of fingerprint image frame, wherein said each determined centroid window is bounded by the corresponding said leading edge column and the corresponding said trailing edge column.

4. The apparatus of claim 3, further comprising a communications interface signal coupled to said computer system.

5. The apparatus of claim 3, wherein said computer system comprises:

a centroid window determiner module.

6. The apparatus of claim 3, wherein said rolled fingerprint capture module comprises:

a pixel knitting module that compares the intensity of each pixel of said each determined centroid window to the intensity of a corresponding pixel of said composite fingerprint image and replaces said corresponding pixel of said composite fingerprint image with said pixel of said each determined centroid window if said pixel of said each determined centroid window is darker than said corresponding pixel of said composite fingerprint image.

7. The apparatus of claim 3, wherein said computer system comprises:

a fingerprint roll detector module.

8. The apparatus of claim 7, wherein said fingerprint roll detector module comprises:

a fingerprint roll start detector module; and

a fingerprint roll stop detector module.

9. The apparatus of claim 3, wherein said rolled fingerprint capture module comprises:

a fingerprint capture module;

a fingerprint image format module coupled to said fingerprint capture module; and

a fingerprint image display module coupled to said fingerprint image format module.

10. A system for rolled fingerprint capture, comprising:

a fingerprint roll start detector module that detects the start of a fingerprint roll in a fingerprint image capturing area;

a centroid window determiner module that determines a centroid window corresponding to each of a plurality of captured fingerprint image frames;

a pixel knitting module that knits pixels of said each determined centroid window into a composite fingerprint image; and

a fingerprint roll stop detector that detects the end of the fingerprint roll;

wherein said centroid window determiner module generates a pixel window in said each of a plurality of captured fingerprint image frames;

wherein said centroid window determiner module comprises a histogram builder;

wherein for each said generated pixel window, said histogram builder builds a histogram representative of the cumulative pixel intensity of pixels present in each column of said generated pixel window;

wherein for each said histogram, said centroid window determiner module scans through said histogram in the direction opposite that in which the finger is rolled for a difference in the total pixel intensities of first adjacent columns of the generated pixel window that is greater than a first fingerprint edge threshold to find a leading edge column;

wherein for each said histogram, said centroid window determiner module scans through said histogram in the direction in which the finger is rolled for a difference in the total pixel intensities of second adjacent columns of the generated pixel window that is greater than a second fingerprint edge threshold to find a trailing edge column; and

wherein said centroid window determiner module generates said determined centroid window in said each of a plurality of fingerprint image frames, wherein said each determined centroid window is bounded by the corresponding said leading edge column and the corresponding said trailing edge column.

11. The system of claim 10, wherein said pixel knitting module knits pixels of each of said determined centroid window into said composite fingerprint image, wherein said pixel knitting module compares the intensity of each pixel of said each of said determined centroid window to the intensity of a corresponding pixel of said composite fingerprint image and replaces said corresponding pixel of said composite fingerprint image with said pixel of said each of said determined centroid window if said pixel of said each of said determined centroid window is darker than said corresponding pixel of said composite fingerprint page.

12. A system for rolled fingerprint capture, comprising:

means for detecting the start of a fingerprint roll;

means for capturing a plurality of fingerprint image frames;

means for determining a centroid window corresponding to each of the plurality of captured fingerprint image frames;

means for knitting pixels of each determined centroid window into a composite fingerprint image; and

means for detecting the end of said fingerprint roll;

wherein said determining means includes;

means for generating a pixel window in a captured fingerprint image frame,

means for finding a leading edge column and a trailing edge column of a fingerprint image in the corresponding generated pixel window, and

means for generating the centroid window in the captured fingerprint image frame, wherein the centroid window is bounded by the leading edge column found and the trailing edge column found;

wherein said edge finding means comprises;

means for building a histogram representative of the cumulative pixel intensity of pixels present in each column of the generated window,

means for scanning through the histogram in the direction opposite that in which the finger is rolled for a difference in the total pixel intensities of first adjacent columns of the generated pixel window that is greater than a first fingerprint edge threshold to find a leading edge, and

means for scanning through the histogram in the direction in which the finger is rolled for a difference in the total pixel intensities of second adjacent columns of the generated pixel window that is greater than a second fingerprint edge threshold to find a trailing edge.

13. A computer program product comprising a computer useable medium having computer program logic recorded thereon for enabling a processor in a computer system to permit a user to capture a rolled fingerprint, said computer program logic comprising:

means for enabling the processor to detect the start of a fingerprint roll;

means for enabling the processor to capture a plurality of fingerprint image frames;

means for enabling the processor to determine a centroid window corresponding to each of the plurality of captured fingerprint image frames;

means for enabling the processor to knit pixels of each determined centroid window into a composite fingerprint image; and

means for enabling the processor to detect the end of said fingerprint roll;

wherein said means for enabling the processor to determine a centroid window comprises;

means for enabling a processor to generate a pixel window in a captured fingerprint image frame,

means for enabling a processor to find a leading edge column and a trailing edge column of a fingerprint image in the corresponding generated pixel window, and

means for enabling a processor to generate the centroid window in the captured fingerprint image frame, wherein the centroid window is bounded by the leading edge column found and the trailing edge column found;

wherein said means for enabling a processor to find a leading edge column and a trailing edge column comprises;

means for enabling a processor to build a histogram representative of the cumulative pixel intensity of pixels present in each column of the generated window,

means for enabling a processor to scan through the histogram in the direction opposite that in which the finger is rolled for a difference in the total pixel intensities of first adjacent columns of the generated pixel window that is greater than a first fingerprint edge threshold to find a leading edge, and

means for enabling a processor to scan through the histogram in the direction in which the finger is rolled for a difference in the total pixel intensities of second adjacent columns of the generated pixel window that is greater than a second fingerprint edge threshold to find a trailing edge.