Comparative texture analysis of tissue for biometric spoof detection

• US 20060062438A1
• Filed: 09/01/2005
• Published: 03/23/2006
• Est. Priority Date: 04/04/2003
• Status: Active Grant

• Current Assignee: HID Global Corporation (Assa Abloy AB)
• Original Assignee: Lumidigm Incorporated (Assa Abloy AB)
• Inventors: Rowe, Robert K.
• Granted Patent: US 7,668,350 B2
• Time in Patent Office: Days
• US Class Current: 382/124
• CPC Class Codes: G06V 40/1324 by using geometrical optics, e.g. using prisms G06V40/1312 takes precedence
• CPC Class Codes: G06V 40/1388 using image processing

Abstract

Methods are described of evaluating the genuineness of a sample presented for biometric evaluation. The sample is illuminated under distinct optical conditions. Light scattered from the sample is received. Multiple images are formed, each image being formed from the received light for one of the optical conditions. A set of texture measures is generated, each texture measure being generated from one of the images. It is determined whether the generated texture measures is consistent with the sample being authentic unconcealed biological tissue.

Claims

1. A method of evaluating genuineness of a sample presented for biometric evaluation, the method comprising:

illuminating the sample under a plurality of distinct optical conditions;

receiving light scattered from the sample for each of the plurality of distinct optical conditions;

forming a plurality of images, each such image being formed from the received light for a respective one of the plurality of distinct optical conditions;

generating a plurality of texture measures, each such texture measure being generated from a respective one of the plurality of images; and

determining whether the generated plurality of texture measures is consistent with the sample being authentic unconcealed biological tissue.

2. The method recited in claim 1 wherein the texture measure comprises an image-contrast measure.

3. The method recited in claim 1 wherein illuminating the sample under a plurality of distinct optical conditions comprises illuminating the sample with light having a plurality of different wavelengths.

4. The method recited in claim 1 wherein illuminating the sample under a plurality of distinct optical conditions comprises illuminating the sample with light under a plurality of distinct polarization conditions.

5. The method recited in claim 1 wherein illuminating the sample under a plurality of distinct optical conditions comprises illuminating the sample with light at a plurality of distinct illumination angles.

6. The method recited in claim 1 wherein generating the plurality of texture measures comprises performing a spatial moving-window analysis of each of the plurality of images.

7. The method recited in claim 6 wherein each of the plurality of texture measures comprises a measure of image contrast within certain spatial frequencies.

8. The method recited in claim 6 wherein:

illuminating the sample under a plurality of distinct optical conditions comprises illuminating the sample with light having a plurality of distinct wavelengths; and

determining whether the generated plurality of texture measures is consistent with the sample being unconcealed biological tissue comprises confirming that an image contrast under red illumination is less than an image contrast under blue illumination.

9. The method recited in claim 6 wherein:

illuminating the sample under a plurality of distinct optical conditions comprises illuminating the sample with light having a plurality of distinct wavelengths; and

determining whether the generated plurality of texture measures is consistent with the sample being unconcealed biological tissue comprises confirming that an image contrast under red illumination is less than a predetermined value.

10. The method recited in claim 6 wherein performing the spatial moving-window analysis comprises calculating moving-window Fourier transforms on the plurality of images.

11. The method received in claim 6 wherein performing the spatial moving-window analysis comprises calculating a moving-window centrality measure and a moving-window variability measure of the plurality of images.

12. The method recited in claim 11 wherein the moving-window centrality measure comprise a moving-window mean and the moving-window variability measure comprises a moving-window standard deviation.

13. The method recited in claim 11 wherein the moving-window centrality measure comprises a moving-window mean and the moving-window variability measure comprises a moving-window range.

14. The method recited in claim 6 wherein:

generating the plurality of texture measures further comprises applying a multidimensional scaling to map the plurality of texture measures to a point in a multidimensional space; and

determining whether the generated plurality of texture measures is consistent with the sample being authentic unconcealed biological tissue comprises determining whether the point is in a predefined region of the multidimensional space.

15. The method recited in claim 14 wherein the multidimensional space is a two-dimensional space.

16. A method of evaluating genuineness of a sample presented for biometric evaluation, the method comprising:

illuminating the sample with light having a plurality of different wavelengths;

receiving light scattered from the sample for each of the different wavelengths;

forming a plurality of images, each such image being formed from the received light for a respective one of the different wavelengths;

generating a plurality of image contrast measures, each such image contrast measure being generated from one of the plurality of images by performing a spatial moving-window analysis of the one of the plurality of images; and

determining from the plurality of image contrast measures whether the sample has an image contrast under red illumination that is less than an image contrast under blue illumination.

17. The method recited in claim 16 wherein performing the spatial moving-window analysis comprises calculating a moving-window Fourier transform on the one of the plurality of images.

18. The method recited in claim 16 wherein performing the spatial moving-window analysis comprises calculating a moving-window centrality measure and a moving-window variability measure of the one of the plurality of images.

19. The method recited in claim 16 wherein:

determining from the plurality of image contrast measures whether the sample has an image contrast under red illumination less than an image contrast under blue illumination comprises applying a multidimensional scaling to map the plurality of image contrast measures to a point in a multidimensional space; and

determining whether the point is in a predefined region of the multidimensional space.

20. An apparatus for evaluating genuineness of a sample presented for biometric evaluation, the apparatus comprising:

means for illuminating the sample under a plurality of distinct optical conditions;

means for receiving light scattered from the sample for each of the plurality of distinct optical conditions;

means for forming a plurality of images, each such image being formed from the received light for a respective one of the plurality of distinct optical conditions;

means for generating a plurality of texture measures, each such texture measure being generated from a respective one of the plurality of images; and

means for determining whether the generated plurality of texture measures is consistent with the sample being authentic unconcealed biological tissue.

21. The apparatus recited in claim 20 wherein the means for generating the plurality of texture measures comprises means for performing a spatial moving-window analysis of each of the plurality of images.

22. The apparatus recited in claim 21 wherein:

the means for generating the optical contrast measure further comprises means for generating a multidimensional scaling to map the figure of merit to a point in a multidimensional space; and

the means for determining whether the generated plurality of texture measures is consistent with the sample being authentic unconcealed biological tissue comprises means for determining whether the point is in a predefined region of the multidimensional space.