BIOMETRIC OBJECT SPOOF DETECTION BASED ON IMAGE INTENSITY VARIATIONS

• US 20180349721A1
• Filed: 02/21/2018
• Published: 12/06/2018
• Est. Priority Date: 06/06/2017
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: AGRAWAL, Mudit, TSUKADA, Akihiro
• Granted Patent: US 10,657,401 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: G06F 2221/2133 Verifying human interaction, e.g., Captcha
• CPC Class Codes: G06V 40/1388 using image processing
• CPC Class Codes: G06V 40/19 Sensors therefor
• CPC Class Codes: G06V 40/197 Matching; Classification
• CPC Class Codes: G06V 40/45 Detection of the body part being alive

Abstract

Apparatus and methods of biometric object spoof detection are configured to receive at least first and second images, including a biometric object, respectively captured at a first and second time in response to a first and second incident light. The first and second incident light is emitted from at least one light source at substantially a same wavelength, but with different sets of illumination characteristics. Further, the apparatus and method are configured to respectively determine a first set and a corresponding second set of reflection intensity features respectively based on at least a part of the first and second images, and to determine a set of reflection intensity difference features based on an intensity difference therebetween. Additionally, the apparatus and methods are configured to classify the biometric object as being a fake object or a real object based on at least one of the reflection intensity difference features.

Claims

1. A method of biometric object spoof detection, comprising:

receiving a first image, including a biometric object, captured at a first time in response to a first incident light emitted from a light source at a wavelength with a first set of illumination characteristics;

receiving a second image, including the biometric object, captured at a second time in response to a second incident light emitted from the light source at the wavelength with a second set of illumination characteristics;

determining a first set of one or more reflection intensity features based on at least a part of the first image;

determining a second set of one or more reflection intensity features based on at least a part of the second image, wherein each of the second set of one or more reflection intensity value features corresponds to one of the first set of one or more reflection intensity features;

determining a set of one or more reflection intensity difference features based on an intensity difference between at least one of the first set of one or more reflection intensity features and a corresponding at least one of the second set of one or more reflection intensity features; and

classifying the biometric object as being a fake object or a real object based on at least one of the set of one or more reflection intensity difference features.

2. The method of claim 1,

wherein the biometric object comprises an eye including a real eye or a fake eye;

wherein determining the set of one or more reflection intensity difference features includes determining a glint intensity difference feature between corresponding corneal glints in the first image and the second image; and

wherein classifying the biometric object as being the fake object or the real object is based on the glint intensity difference feature.

3. The method of claim 1,

wherein determining the set of one or more reflection intensity difference features includes determining an overall image intensity difference feature between the first image and the second image;

and

wherein classifying the biometric object as being the fake object or the real object is based on the overall image intensity difference feature.

4. The method of claim 1, further comprising:

wherein the biometric object comprises an eye;

wherein determining the set of one or more intensity difference features includes determining one or more eye-part intensity difference features between corresponding eye-parts in the first image and the second image; and

wherein classifying the biometric object as being the fake object or the real object is based on the one or more eye-part intensity difference features.

5. The method of claim 4, wherein the corresponding eye-parts in the first image and the second image include at least one of a pupil, an iris, a sclera, or a periocular region of the first image and the second image.

6. The method of claim 1, wherein determining the set of one or more reflection intensity difference features includes determining one or more of an intensity amplitude-based metric, an intensity variance-based metric, a local binary pattern-based metric, a pixel intensity-based metric, a local texture-based metric, or an intensity feature-based histogram.

7. The method of claim 1, wherein the first set of illumination characteristics and the second set of illumination characteristics each comprise a different value of at least one of an illumination power, an exposure time, an angle of incidence, or any characteristic resulting in a different signal-to-noise ratio between the first image and the second image.

8. The method of claim 1, wherein the wavelength of the light source comprises one of a plurality of wavelengths within an infrared band of light.

9. The method of claim 1, further comprising:

wherein the biometric object comprises an eye; and

capturing the first image and the second image at a frame rate such that a position of at least a portion of the biometric object is substantially similar between the first image and the second image.

10. The method of claim 2, further comprising:

detecting two or more glint-like objects in at least one of the first image and the second image;

determining a size and a shape, respectively, of the two or more glint-like objects;

distinguishing at least one glint from other glass-reflection-based glare based on the size and the shape, respectively, of the two or more glint-like objects.

11. The method of claim 1, further comprising:

performing biometric identification based on the first image or the second image, or based on one or more additional images of the biometric object received in response to an additional incident light emitted from the light source at the wavelength.

12. A computer device for biometric object spoof detection, comprising:

a processor; and

a memory in communication with the processor;

wherein the processor is configured to;

receive a first image, including a biometric object, captured at a first time in response to a first incident light emitted from a light source at a wavelength with a first set of illumination characteristics;

receive a second image, including the biometric object, captured at a second time in response to a second incident light emitted from the light source at the wavelength with a second set of illumination characteristics;

determine a first set of one or more reflection intensity features based on at least a part of the first image;

determine a second set of one or more reflection intensity features based on at least a part of the second image, wherein each of the second set of one or more reflection intensity features corresponds to one of the first set of one or more reflection intensity features;

determine a set of one or more intensity difference features based on an intensity difference between at least one of the first set of one or more reflection intensity features and a corresponding at least one of the second set of one or more reflection intensity features; and

classify the biometric object as being a fake object or a real object based on at least one of the set of one or more reflection intensity difference features.

13. The device of claim 12, wherein the biometric object comprises an eye including a real eye or a fake eye, and wherein the processor is further configured to:

determine a glint intensity difference feature between corresponding corneal glints in the first image and the second image; and

classify the biometric object as being the fake object or the real object based on the glint intensity difference feature.

14. The device of claim 12, wherein the processor is further configured to:

determine an overall image intensity difference feature between the first image and the second image; and

classify the biometric object as being the fake object or the real object based on the overall image intensity difference feature.

15. The device of claim 12, wherein the biometric object comprises an eye, and wherein the processor is further configured to:

determine one or more eye-part intensity difference features between corresponding eye-parts in the first image and the second image; and

classify the biometric object as being the fake object or the real object based on the one or more eye-part intensity difference features.

16. The device of claim 15, wherein the corresponding eye-parts in the first image and the second image include at least one of a pupil, an iris, a sclera, or a periocular region of the first image and the second image.

17. The device of claim 12, wherein the set of one or more reflection intensity difference features includes one or more of an intensity amplitude-based metric, an intensity variance-based metric, a local binary pattern-based metric, a pixel intensity-based metric, a local texture-based metric, or an intensity feature-based histogram.

18. The device of claim 12, wherein the first set of illumination characteristics and the second set of illumination characteristics each comprise a different value of at least one of an illumination power, an exposure time, an angle of incidence, or any characteristic resulting in a different signal-to-noise ratio between the first image and the second image.

19. The device of claim 12, wherein the wavelength of the light source comprises one of a plurality of wavelengths within an infrared band of light.

20. The device of claim 12, wherein the biometric object comprises an eye, and wherein the processor is further configured to:

capture the first image and the second image at a frame rate such that a position of at least a portion of the biometric object is substantially similar between the first image and the second image.

21. The device of claim 13, wherein the processor is further configured to:

detect two or more glint-like objects in at least one of the first image and the second image;

determine a size and a shape, respectively, of the two or more glint-like objects; and

distinguish at least one glint from other glass-reflection-based glare based on the size and the shape, respectively, of the two or more glint-like objects.

22. The device of claim 12, wherein the processor is further configured to:

perform biometric identification based on the first image or the second image, or based on one or more additional images of the biometric object received in response to an additional incident light emitted from the light source at the wavelength.

23. A computer readable medium storing one or more instructions executable by a computer device, the instructions causing the computer device to:

receive a first image, including a biometric object, captured at a first time in response to a first incident light emitted from a light source at a wavelength with a first set of illumination characteristics;

receive a second image, including the biometric object, captured at a second time in response to a second incident light emitted from the light source at the wavelength with a second set of illumination characteristics;

determine a first set of one or more reflection intensity features based on at least a part of the first image;

determine a second set of one or more reflection intensity features based on at least a part of the second image, wherein each of the second set of one or more reflection intensity features corresponds to one of the first set of one or more reflection intensity features;

determine a set of one or more intensity difference features based on an intensity difference between at least one of the first set of one or more reflection intensity features and a corresponding at least one of the second set of one or more reflection intensity features; and

classify the biometric object as being a fake object or a real object based on at least one of the set of one or more reflection intensity difference features.

24. The computer readable medium of claim 23, wherein the biometric object comprises an eye including a real eye or a fake eye, and wherein the instructions further cause the computer device to:

determine a glint intensity difference feature between corresponding corneal glints in the first image and the second image; and

classify the biometric object as being the fake object or the real object based on the glint intensity difference feature.

25. The computer readable medium of claim 23, wherein the instructions further cause the computer device to:

determine an overall image intensity difference feature between the first image and the second image; and

classify the biometric object as being the fake object or the real object based on the overall image intensity difference feature.

26. The computer readable medium of claim 23, wherein the biometric object comprises an eye, and wherein the instructions further cause the computer device to:

determine one or more eye-part intensity difference features between corresponding eye-parts in the first image and the second image; and

classify the biometric object as being the fake object or the real object based on the one or more eye-part intensity difference features.