System and method for three-dimensional biometric data feature detection and recognition

• US 20110216948A1
• Filed: 03/04/2010
• Published: 09/08/2011
• Est. Priority Date: 03/04/2010
• Status: Active Grant

• Current Assignee: Flashscan3D, LLC
• Original Assignee: Flashscan3D, LLC
• Inventors: Fleming, Robert Kyle, Boles, Colby, Yalla, Veeraganesh, Daley, Raymond Charles
• Granted Patent: US 8,391,590 B2
• Time in Patent Office: Days
• US Class Current: 382/125
• CPC Class Codes: G06V 20/64 Three-dimensional objects
• CPC Class Codes: G06V 20/653 by matching three-dimensional models, e.g. conformal mapping of Riemann surfaces
• CPC Class Codes: G06V 40/10 Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
• CPC Class Codes: G06V 40/12 Fingerprints or palmprints
• CPC Class Codes: G06V 40/1312 direct reading, e.g. contactless acquisition

Abstract

The system includes a 3D feature detection module and 3D recognition module 202. The 3D feature detection module processes 3D surface map of a biometric object, wherein the 3D surface map includes a plurality of 3D coordinates. The 3D feature detection module determines whether one or more types of 3D features are present in the 3D surface map and generates 3D feature data including 3D coordinates and feature type for the detected features. The 3D recognition module compares the 3D feature data with biometric data sets for identified persons. The 3D recognition module determines a match between the 3D feature data and one of the biometric data sets when a confidence value exceeds a threshold.

Claims

1. A method for three-dimensional (3D) feature detection, comprising:

receiving a 3D surface map, wherein the 3D surface map includes a plurality of points having 3D coordinates;

generating one or more subsets of the plurality of points of the 3D surface map;

processing the 3D surface map to detect a 3D feature within one or more of the subsets;

generating 3D feature data for one or more of the 3D features detected in the 3D surface map, wherein the 3D feature data includes 3D location data.

2. The method of claim 1, wherein the 3D location data includes 3D coordinates.

3. The method of claim 1, wherein the 3D location data includes 3D coordinates and orientation angles of the 3D feature.

4. The method of claim 1, wherein the detecting step of whether a 3D feature is present within the subset of plurality of points, comprises:

comparing the subset of the plurality of points to a 3D feature template, wherein the 3D feature template has an associated predetermined 3D feature;

assigning a matching value in response to a probability that the subset of the plurality of points matches the 3D feature template; and

when the probability value exceeds a predetermined threshold, determining that the subset includes the predetermined 3D feature associated with the 3D feature template.

5. The method of claim 1, wherein the detecting step of whether a 3D feature is present within the subset of plurality of points, comprises:

comparing the subset with a plurality of 3D feature templates, wherein each of the 3D feature templates is associated with a different predetermined 3D feature;

assigning a matching value to each comparison of the plurality of 3D feature templates with the subset;

determining the comparison with a most favorable matching value; and

determining the subset includes the predetermined 3D feature with the most favorable matching value.

6. The method of claim 5, wherein the 3D feature template is derived from a plurality of training sets, wherein each training set includes a known plurality of points encoded with the 3D feature and wherein an analysis of the plurality of training sets is performed to generate the 3D feature template using at least one of:

a PCA based algorithm, ICA based algorithm, LDA based algorithm, Kernel-PCA based algorithm, SVM algorithm and neural network algorithm.

7. The method of claim 1, wherein 3D biometric image data includes 3D image data of a fingerprint and wherein the 3D feature template is associated with at least one of the following predetermined 3D features:

a ridge shape, a ridge height, a valley depth, a valley shape, a ridge flow gradient, ridge height gradient, finger curvature, finger shape and finger size.

8. The method of claim 7, wherein the 3D feature template is further associated with at least one of the following predetermined classifications:

a whorl, an arc, a loop, a delta, a ridge ending, ridge orientation, a type of minutia point.

9. The method of claim 5, wherein the step of comparing the subset of points with a plurality of 3D feature templates, wherein each of the 3D feature templates is associated with a different predetermined classification, comprises:

performing a comparison of the subset of points with each of the plurality of 3D feature templates using a trained feature classifier algorithm.

10. The method of claim 9, wherein the trained feature classifier algorithm includes at least one of:

a PCA based algorithm, ICA based algorithm, LDA based algorithm, Kernel-PCA based algorithm, SVM algorithm and neural network algorithm.

11. The method of claim 1, wherein the step of when a 3D feature is detected within a subset, generating 3D feature data for the 3D feature, wherein the 3D feature data includes 3D location data, further comprises:

when a 3D feature is detected within a subset, processing the plurality of points in the subset for validation of the 3D feature;

assigning a quality assessment value to the 3D feature of the subset.

12. The method of claim 1, further including:

processing one or more images of a 3D object having a projected structured light illumination (SLI) pattern;

receiving calibrations for an SLI system that generated the one or more images of the 3D object;

storing the one or images and the calibrations of the SLI system in a database; and

generating the 3D surface map of the 3D object from the one or more images of the 3D object and the calibrations of the SLI system.

13. A method for three-dimensional (3D) biometric recognition, comprising:

receiving 3D feature data for one or more 3D features detected in a 3D surface map, wherein the 3D feature data includes 3D location data for each of the one or more 3D features;

comparing the 3D feature data with a plurality of biometric 3D data sets, wherein each of the plurality of 3D biometric data sets includes 3D location data for one or more 3D features;

generating a confidence value in response to a comparison between the 3D feature data and one of the plurality of biometric 3D data sets;

determining a match between the 3D feature data and the one of the plurality of 3D biometric data sets when the confidence value exceeds a threshold.

14. The method of claim 13, wherein the 3D location data for each of the one or more 3D features includes 3D coordinates.

15. The method of claim 13, wherein the 3D location data associated with one of the 3D features includes 3D coordinates and orientation angles of the 3D feature and a relative position of the 3D feature with respect to other 3D features detected in the 3D surface map.

16. The method of claim 13, wherein comparing the 3D feature data with a plurality of biometric 3D data sets, includes:

generating a confidence value in response to each comparison between one of the plurality of biometric 3D data sets and the 3D feature data; and

determining a match between the comparison of the 3D feature data and one of the plurality of 3D biometric data sets having a most favorable confidence value and when the confidence value exceeds a threshold.

17. The method of claim 13, wherein comparing the 3D feature data with a plurality of biometric 3D data sets, includes:

determining at least one categorization for the 3D feature data;

determining the plurality of biometric 3D data sets including the categorization; and

comparing the 3D feature data with the plurality of biometric 3D data sets including the categorization.

18. The method of claim 17, wherein comparing the 3D feature data with a plurality of biometric 3D data sets, further comprises:

comparing the 3D features in the 3D feature data to a plurality of 3D features in one of the biometric 3D data sets;

generating a matching value for matching 3D features.

19. The method of claim 18, further comprising:

determining a number of clusters of matching 3D features; and

generating a confidence value based on the 3D matching values and number of clusters of matching 3D features.

20. The method of claim 19, further comprising:

comparing the 3D feature data generated from the 3D surface map with 2D feature data from one or more of the plurality of biometric data sets, wherein the 2D feature data includes 2D coordinates of one or more features associated with the biometric data set;

generating a matching value for matching 2D features; and

generating a confidence value based on the 3D matching values and the 2D matching values.

21. The method of claim 13, wherein the 3D features include:

a ridge shape, a ridge height, a valley depth, a valley shape, a ridge flow gradient, ridge height gradient, finger curvature, finger shape and finger size.

22. A method for three-dimensional (3D) feature detection, comprising:

receiving a 3D surface map of a fingerprint, wherein the 3D surface map includes a plurality of points having 3D coordinates;

generating one or more subsets of the plurality of points of the 3D surface map;

processing the 3D surface map to detect at least one 3D feature within one or more of the subsets, wherein the 3D feature includes at least one of;

a ridge shape, a ridge height, a valley depth, a valley shape, a ridge flow gradient, ridge height gradient;

generating 3D feature data for the 3D feature detected in the 3D surface map, wherein the 3D feature data includes 3D coordinates;

comparing the 3D feature data with a plurality of biometric 3D data sets, wherein each of the plurality of 3D biometric data sets includes 3D coordinates for at least one type of 3D feature;

generating a confidence value in response to a comparison between the 3D feature data and one of the plurality of biometric 3D data sets; and

determining a match between the 3D feature data and the one of the plurality of 3D biometric data sets when the confidence value exceeds a threshold.

23. The method of claim 22, further comprising:

determining a relative distance and orientation angle between features in the 3D surface map;

receiving a relative distance and orientation angle between features based on a 2D binary image; and

comparing the relative distance and orientation angle between features in the 3D surface map with the relative distance and orientation angle between features based on the 2D binary image; and

generating a confidence value based on the comparison.