Finger-based identification systems and methods

• US 8,594,391 B2
• Filed: 05/06/2004
• Issued: 11/26/2013
• Est. Priority Date: 05/06/2004
• Status: Active Grant

• Current Assignee: PixArt Imaging Incorporated
• Original Assignee: Avago Technologies General IP PTE Limited (Broadcom, Inc.)
• Inventors: Machida, Akihiro
• Primary Examiner(s): Park, Chan S
• Assistant Examiner(s): YEH, EUENG NAN
• Application Number:
  US10/839,921
• Publication Number:
  US 20050249387A1
• Time in Patent Office: 3,491 Days
• Field of Search: 382/115, 382/119, 382/124
• US Class Current: 382/124
• CPC Class Codes: G06V 40/12 Fingerprints or palmprints
• CPC Class Codes: G06V 40/20 Movements or behaviour, e.g. gesture recognition recognition of facial expressions G06V40/16

Abstract

Finger-based identification systems and methods are described. In one aspect, a two-dimensional fingerprint image of a person'"'"'s finger contacting an input surface is captured. Finger trace data is captured from movement of the person'"'"'s finger over the input surface. Based on the captured two-dimensional fingerprint image and the captured finger trace data, a determination is made whether the person corresponds to an enrolled person.

Claims

1. An identification method, comprising:

causing an identification system to enter a fingerprint capturing mode;

while the identification system is in the fingerprint capturing mode, capturing a two-dimensional fingerprint image of a person'"'"'s finger contacting an input surface;

while the identification system is in the fingerprint capturing mode, converting the two-dimensional fingerprint image into candidate fingerprint data;

while the identification system is in the fingerprint capturing mode, comparing the candidate fingerprint data with each reference fingerprint data in a set of reference fingerprint data;

based on the comparison of the candidate fingerprint data with the set of reference fingerprint data, identifying a first reference fingerprint data that matches the candidate fingerprint data more than any other reference fingerprint data in the set of reference fingerprint data, the first reference fingerprint data corresponding to reference fingerprint data of a first enrolled person;

after identifying the first reference fingerprint data, causing the identification system to enter a navigation mode;

while the identification system is in the navigation mode, capturing finger trace data of a movement of the person'"'"'s finger relative to the input surface as the person'"'"'s finger moves relative to the input surface;

converting the finger trace data into a sequence of directional symbols to form a candidate trace data passcode;

mapping the candidate trace data passcode into a candidate personal identifier;

obtaining a reference personal identifier associated with the first enrolled person identified during the fingerprint capturing mode;

comparing the candidate personal identifier with the reference personal identifier; and

determining that the person corresponds to the first enrolled person only if the candidate personal identifier matches the reference personal identifier.

2. The method of claim 1, wherein the two-dimensional image is captured in portions during a period of relative movement between the person'"'"'s finger and the input surface.

3. The method of claim 1, wherein the identification system does not operate simultaneously in the fingerprint capturing mode and the navigation mode.

4. The method of claim 3, wherein a similarity between the candidate fingerprint data and each reference fingerprint data in the set of reference fingerprint data is computed using at least one of a pattern matching technique and a graph-theory-based matching technique and wherein the first reference fingerprint data comprises a similarity that at least one of meets and exceeds a threshold measure of similarity to define a match between the candidate fingerprint data and the first reference fingerprint data.

5. The method of claim 1, wherein converting the captured finger trace data comprises computing directional data describing movement directions of the person'"'"'s finger over the input surface.

6. The method of claim 1, wherein capturing finger trace data comprises capturing multiple images of the person'"'"'s finger moving over the input surface.

7. The method of claim 6, further comprising tracking fingerprint features across the multiple images, wherein tracking fingerprint features comprises correlating fingerprint features across the multiple images.

8. The method of claim 1, wherein the set of reference fingerprint data at least includes second reference fingerprint data corresponding to a reference fingerprint data of a second enrolled person in addition to comprising the first reference fingerprint data.

9. The method of claim 1, wherein the first reference fingerprint data includes at least one of singularities extracted from the two-dimensional fingerprint image, minutia extracted from the two-dimensional fingerprint image, and characteristic patterns extracted from the two-dimensional fingerprint image.

10. The method of claim 1, further comprising enrolling a person by:

capturing a two-dimensional fingerprint image of a person'"'"'s finger contacting an input surface;

computing reference fingerprint data from the captured two-dimensional fingerprint image;

capturing finger trace data from movement of the person'"'"'s finger over the input surface;

mapping the captured finger trace data to a candidate personal identifier; and

storing the reference fingerprint data and the candidate personal identifier.

21. The method of claim 1, wherein the capturing of the finger trace data comprises capturing the finger trace data while the person'"'"'s finger is moving in contact with the input surface.

22. The method of claim 1, wherein the first reference fingerprint data includes at least one of singularities extracted from the two-dimensional fingerprint image, minutia extracted from the two-dimensional fingerprint image, and characteristic patterns extracted from the two-dimensional fingerprint image.

11. An identification system, comprising:

an input surface;

a sensor operable to capture a two-dimensional fingerprint image of a person'"'"'s finger contacting an input surface during a capturing mode and to capture finger trace data of a movement of the person'"'"'s finger relative to the input surface as the person'"'"'s finger moves relative to the input surface during a finger navigation mode;

a processing system operable switch between operating in the capturing mode and the navigation mode, the processing system being further operable, while in the capturing mode, to perform the following;

convert the two-dimensional fingerprint image into candidate fingerprint data;

compare the candidate fingerprint data with each reference fingerprint data in a set of reference fingerprint data; and

based on the comparison of the candidate fingerprint data with the set of reference fingerprint data, identify a first reference fingerprint data that matches the candidate fingerprint data better than any other reference fingerprint data in the set of reference fingerprint data, the first reference fingerprint data corresponding to reference fingerprint data of a first enrolled person;

the processing system being further operable, while in the navigation mode, to perform the following;

convert the finger trace data into a sequence of directional symbols to form a candidate trace data passcode;

map the candidate trace data passcode into a candidate personal identifier;

obtain a reference personal identifier associated with the first enrolled person identified during the fingerprint capturing mode;

compare the candidate personal identifier with the reference personal identifier; and

determine that the person corresponds to the first enrolled person only if the candidate personal identifier matches the reference personal identifier.

12. The system of claim 11, wherein the sensor is operable to capture the two-dimensional fingerprint image in portions during a period of relative movement between the person'"'"'s finger and the input surface.

13. The system of claim 11, wherein the processing system does not operate simultaneously in the capturing mode and the navigation mode.

14. The system of claim 13, wherein a similarity between the candidate fingerprint data and each reference fingerprint data in the set of reference fingerprint data is computed using at least one of a pattern matching technique and a graph-theory-based matching technique and wherein the first reference fingerprint data comprises a similarity that at least one of meets and exceeds a threshold measure of similarity to define a match between the candidate fingerprint data and the first reference fingerprint data.

15. The system of claim 13, wherein the processing system is operable to compute directional data describing movement directions of the person'"'"'s finger over the input surface.

16. The system of claim 11, wherein the sensor is operable to capture multiple images of the person'"'"'s finger moving over the input surface.

17. The system of claim 16, wherein the processing system is operable to track fingerprint features across the multiple images.

18. The system of claim 17, wherein the processing system is operable to track fingerprint features by correlating fingerprint features across the multiple images.

19. The system of claim 11, wherein the processing system is operable to match the captured two-dimensional fingerprint image to reference fingerprint data obtained from at least one enrolled person.

20. The system of claim 11, wherein the processing system is operable to enroll a person by:

computing reference fingerprint data from a two-dimensional fingerprint image of a person'"'"'s finger contacting the input surface;

mapping finger trace data captured from movement of the person'"'"'s finger over the input surface to a candidate personal identifier; and

storing the reference fingerprint data and the candidate personal identifier.