Technologies for efficient identity recognition based on skin features

• US 10,181,073 B2
• Filed: 06/29/2016
• Issued: 01/15/2019
• Est. Priority Date: 06/29/2016
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Chattopadhyay, Rita, Alam, Tina
• Primary Examiner(s): Gilliard, Delomia L
• Application Number:
  US15/196,169
• Publication Number:
  US 20180005019A1
• Time in Patent Office: 930 Days
• Field of Search: None
• CPC Class Codes: G06F 18/2411 based on the proximity to a decision surface, e.g. support vector machines
• CPC Class Codes: G06V 10/50 by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
• CPC Class Codes: G06V 10/56 relating to colour
• CPC Class Codes: G06V 40/1347 Preprocessing; Feature extraction
• CPC Class Codes: G06V 40/171 Local features and components; Facial parts eye characteristics G06V40/18; Occluding parts, e.g. glasses; Geometrical relationships
• CPC Class Codes: G06V 40/172 Classification, e.g. identification

Abstract

Technologies for efficient identity recognition based on skin features include a compute device. The compute device includes an image capture device and an image acquisition module to obtain, with the image capture device, an image that depicts the skin of the person. The compute device also includes a skin feature determination module to identify pixels in the obtained image that are associated with the skin of the person, and determine one or more features of the skin based on the identified pixels. Additionally, the compute device includes an identity determination module to generate a feature vector that includes the determined features of the skin, and analyze the feature vector with reference data to determine an identity of the person. Other embodiments are described and claimed.

Claims

1. A compute device to identify a person based on skin features, the compute device comprising:

an image capture device;

an image acquisition module to obtain, with the image capture device, an image that depicts the skin of the person;

a skin feature determination module to (i) identify, in the obtained image, pixels that have a color component that satisfies a threshold value defined as a function of a relationship between other color components of the same pixel as pixels that are associated with the skin of the person, and (ii) determine one or more features of the skin based on the identified pixels; and

an identity determination module to (i) generate a feature vector that includes the determined features of the skin, and (ii) analyze the feature vector with reference data to determine an identity of the person.

2. The compute device of claim 1, wherein to determine one or more features of the skin based on the identified pixels comprise to determine the one or more features of the skin based on a color of each identified pixel.

3. The compute device of claim 1, wherein to determine the one or more features comprises to:

determine a mode for each of red, green, blue, and gray values of the identified pixels; and

determine an average of each of the red, green, and blue values.

4. The compute device of claim 3, wherein to determine the one or more features further comprises to determine normalized values of the averages of the red, green, and blue values.

5. The compute device of claim 1, wherein to analyze the feature vector comprises to provide the feature vector to a support vector machine (SVM) classifier with a pre-trained SVM model that is based on other images that depict skin.

6. The compute device of claim 1, wherein to obtain the image comprises to obtain an image of the face of the person.

7. The compute device of claim 1, wherein to obtain the image comprises to obtain an image of a palm of the person.

8. The compute device of claim 1, wherein the identity determination module is further to output an indication of the determined identity and a confidence value indicative of a confidence that the determined identity is accurate.

9. The compute device of claim 1, further comprising an eye distance determination module to determine a distance between the eyes of the person based on the obtained image, wherein the identity determination module is further to add the determined distance to the feature vector to determine the identity of the person based additionally on the determined distance.

10. The compute device of claim 1, wherein the skin feature determination module is further to set, for each pixel that is not identified as a pixel associated with the skin, one or more color values of the pixel to a predefined value indicative that the corresponding pixel is not representative of skin.

11. One or more non-transitory computer-readable storage media comprising a plurality of instructions that, when executed by a compute device, cause the compute device to:

obtain, with an the image capture device of the compute device, an image that depicts the skin of a person;

identify, in the obtained image, pixels that have a color component that satisfies a threshold value defined as a function of a relationship between other color components of the same pixel as pixels that are associated with the skin of the person;

determine one or more features of the skin based on the identified pixels;

generate a feature vector that includes the determined features of the skin; and

analyze the feature vector with reference data to determine an identity of the person.

12. The one or more non-transitory computer-readable storage media of claim 11, wherein to determine one or more features of the skin based on the identified pixels comprises to determine the one or more features of the skin based on a color of each identified pixel.

13. The one or more non-transitory computer-readable storage media of claim 11, wherein to determine the one or more features comprises to:

determine a mode for each of red, green, blue, and gray values of the identified pixels; and

determine an average of each of the red, green, and blue values.

14. The one or more non-transitory computer-readable storage media of claim 13, wherein to determine the one or more features further comprises to determine normalized values of the averages of the red, green, and blue values.

15. The one or more non-transitory computer-readable storage media of claim 11, wherein to analyze the feature vector comprises to provide the feature vector to a support vector machine (SVM) classifier with a pre-trained SVM model that is based on other images that depict skin.

16. The one or more non-transitory computer-readable storage media of claim 11, wherein to obtain the image comprises to obtain an image of the face of the person.

17. The one or more non-transitory computer-readable storage media of claim 11, wherein to obtain the image comprises to obtain an image of a palm of the person.

18. The one or more non-transitory computer-readable storage media of claim 11, wherein the plurality of instructions further cause the compute device to output an indication of the determined identity and a confidence value indicative of a confidence that the determined identity is accurate.

19. The one or more non-transitory computer-readable storage media of claim 11, wherein the plurality of instructions further cause the compute device to:

determine a distance between the eyes of the person based on the obtained image; and

add the determined distance to the feature vector to determine the identity of the person based additionally on the determined distance.

20. A method to identify a person based on skin features, comprising:

obtaining, with an image capture device of a compute device, an image that depicts the skin of the person;

identifying, by the compute device and in the obtained image, pixels that have a color component that satisfies a threshold value defined as a function of a relationship between other color components of the same pixel as pixels that are associated with the skin of the person;

determining, by the compute device, one or more features of the skin based on the identified pixels;

generating, by the compute device, a feature vector that includes the determined features of the skin; and

analyzing, by the compute device, the feature vector with reference data to determine an identity of the person.

21. The method of claim 20, wherein determining one or more features of the skin based on the identified pixels comprises determining the one or more features of the skin based on a color of each identified pixel.

22. The method of claim 20, wherein determining the one or more features comprises:

determining a mode for each of red, green, blue, and gray values of the identified pixels; and

determining an average of each of the red, green, and blue values.

23. The method of claim 22, wherein determining the one or more features further comprises determining normalized values of the averages of the red, green, and blue values.