3D ENHANCED IMAGE CORRECTION

• US 20160142702A1
• Filed: 03/20/2015
• Published: 05/19/2016
• Est. Priority Date: 11/13/2014
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Ghosh, Rahuldeva, Pfursich, Scott, Graumann, David L.
• Granted Patent: US 10,764,563 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/10004 Still image; Photographic image
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/10028 Range image; Depth image; 3D point clouds
• CPC Class Codes: G06T 2207/20012 Locally adaptive
• CPC Class Codes: G06T 2207/30201 Face
• CPC Class Codes: G06T 5/80 Geometric correction
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06V 10/44 Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
• CPC Class Codes: G06V 20/64 Three-dimensional objects
• CPC Class Codes: G06V 40/172 Classification, e.g. identification
• CPC Class Codes: H04N 13/271 wherein the generated image signals comprise depth maps or disparity maps
• CPC Class Codes: H04N 13/293 Generating mixed stereoscopic images; Generating mixed monoscopic and stereoscopic images, e.g. a stereoscopic image overlay window on a monoscopic image background
• CPC Class Codes: H04N 2013/0092 Image segmentation from stereoscopic image signals
• CPC Class Codes: H04N 2213/003 Aspects relating to the "2D+depth" image format
• CPC Class Codes: H04N 23/70 Circuitry for compensating brightness variation in the scene
• CPC Class Codes: H04N 23/71 Circuitry for evaluating the brightness variation

Abstract

A user authentication system and method. A two-dimensional image of a scene is obtained and range information obtained from the scene is aligned with the two-dimensional image. One or more depth regions is identified and image segments corresponding to the one or more depth regions are selected within the two-dimensional image. Brightness operations are performed on one or more of the selected image segments to form a corrected image.

Claims

1. A system for enhanced image correction, comprising:

a 2D camera;

a 3D camera; and

a processor to;

receive a two-dimensional image from the 2D camera;

receive range information from the 3D camera;

align the range information with the two-dimensional image;

identify one or more depth regions;

select, within the two-dimensional image, image segments corresponding to the one or more depth regions;

perform brightness operations on one or more of the image segments to form a corrected image; and

authenticate a user based on a user feature in the corrected image.

2. The system of claim 1, wherein the two-dimensional image is a video frame.

3. The system of claim 1, wherein to select image segments, the processor is to generate a pixel polygon descriptor for each image segment.

4. The system of claim 1, wherein to align the range information with the two-dimensional image, the processor is to convert the range information to a coordinate system.

5. The system of claim 1, wherein the range information is segmented based on world coordinate contours.

6. The system of claim 1, wherein to select the image segment, the processor is to project the depth region on the two-dimensional image.

7. The system of claim 1, wherein the processor is to determine, for each selected image segment, a desired pixel brightness in the hue saturation brightness space and modifies 2D camera controls as necessary to achieve the desired pixel brightness in the hue saturation brightness space.

8. The system of claim 7, wherein the desired pixel brightness is approximately 128.

9. The system of claim 7, wherein the processor is to modify 2D camera controls to apply to each selected image segment, independent pixel integration times based on the desired pixel brightness.

10. The system of claim 7, wherein the processor is to modify 2D camera controls to apply to each selected image segment, amplification values based on the desired pixel brightness.

11. The system of claim 7, wherein the processor is to search for a user'"'"'s feature in the corrected image and authenticates the user based on the user feature.

12. The system of claim 7, wherein the processor is to search for a face in the corrected image and authenticates a user based on facial features in the image.

13. The system of claim 12, wherein the processor is to authenticate a user by executing a cloud-based authentication routine.

14. A method of enhanced image correction, comprising:

obtaining a two-dimensional image of a scene;

aligning range information obtained from the scene with the two-dimensional image;

identifying one or more depth regions;

selecting, within the two-dimensional image, image segments corresponding to the one or more depth regions; and

performing brightness operations on one or more of the selected image segments to form a corrected image.

15. The method of claim 14, wherein aligning range information obtained from a scene with a two-dimensional image of the scene includes extracting the image from video.

16. The method of claim 14, wherein selecting image segments includes generating a pixel polygon descriptor for each image segment.

17. The method of claim 14, wherein performing the brightness operation comprises:

determining, for each selected image segment, a desired pixel brightness in the hue saturation brightness space; and

applying independent pixel integration times and amplification values as necessary to achieve the desired pixel brightness in the hue saturation brightness space.

18. The method of claim 17, wherein the desired pixel brightness is approximately 128.

19. The method of claim 17, wherein the method further comprises searching for a user'"'"'s feature in the corrected image and authenticating the user based on the user feature.

20. The method of claim 17, wherein the method further comprises searching for a face in the corrected image and authenticating a user based on facial features in the image.

21. At least one machine-readable medium including instructions that, when performed by a machine, cause the machine to:

obtain a two-dimensional image of a scene;

align range information obtained from the scene with the two-dimensional image;

identify one or more depth regions;

select, within the two-dimensional image, image segments corresponding to the one or more depth regions; and

perform brightness operations on one or more of the selected image segments to form a corrected image.

22. The at least one machine-readable medium of claim 21, wherein the instructions to select, within the two-dimensional image, image segments corresponding to the one or more depth regions include instructions to identify contours in the image.

23. The at least one machine-readable medium of claim 22, wherein a contour is a consistently ascending or descending slope with less than a threshold level of variation in slope over a predefined kernel region.

24. The at least one machine-readable medium of claim 22, wherein a contour is a consistently ascending or descending slope with less than 5 mm variation in slope over a 1 cm×

1 cm×

1 cm kernel region.

25. The at least one machine-readable medium of claim 21, wherein the instructions to perform the brightness operation include instructions to distribute the brightness operation for each image segment to a separate processor.