Techniques for disparity estimation using camera arrays for high dynamic range imaging

Patent metadata

Field	Value
Publication number	US-9600887-B2
Application number	US-201314101061-A
Country	US
Kind code	B2
Filing date	Dec 9, 2013
Priority date	Dec 9, 2013
Publication date	Mar 21, 2017
Grant date	Mar 21, 2017

Abstract

Techniques for improved focusing of camera arrays are described. In one embodiment, a system may include a processor circuit, a camera array, and an imaging management module for execution on the processor circuit to capture an array of images from the camera array, the array of images comprising first and second images taken with first and second values of an exposure parameter, respectively, the first value different than the second value, to estimate a noise level, to normalize an intensity of each image based upon the noise level of the respective image, to produce a respective normalized image, to identify candidate disparities in each of the respective normalized images, to estimate a high dynamic range (HDR) image patch for each candidate disparity, and to compute an error from the HDR image patch and an objective function, to produce a disparity estimate. Other embodiments are described and claimed.

First claim

The invention claimed is:

1. At least one non-transitory machine-readable medium comprising a plurality of instructions that, in response to being executed on a computing device, cause the computing device to:
receive an array of images from a camera array, the array of images to comprise a plurality of images including at least a first image taken with a first value of an exposure parameter and a second image taken with a second value of the exposure parameter, the first value different than the second value;
estimate a noise level of each image of the array of images;
generate a plurality of exposure-normalized images, each one of the plurality of exposure-normalized images to be generated by normalizing a respective one of the plurality images based on an exposure parameter value used to capture that one of the plurality of images and a median exposure level for the plurality of images;
identify candidate disparities in each of the plurality of exposure-normalized images;
estimate a high dynamic range (HDR) image patch for each candidate disparity; and
compute an error from the HDR image patch and an objective function to produce a disparity estimate.

2. The at least one non-transitory machine-readable medium of claim 1 , comprising instructions that, in response to being executed on the computing device, cause the computing device to:
produce a composite image having a higher dynamic range than any one of the array of images, by use of the disparity estimate.

3. The at least one non-transitory machine-readable medium of claim 1 , the array of images comprising a first image taken with a first value of ISO gain and a second image taken with a second value of ISO gain.

4. The at least one non-transitory machine-readable medium of claim 1 , the camera array comprising one of a single row of cameras and a single column of cameras.

5. The at least one non-transitory machine-readable medium of claim 1 , estimating the HDR image patch comprising:
designating a reference view as one image of the array of images;
computing a respective disparity between the reference view and each other image of the array of images;
shifting each exposure-normalized image by its respective disparity; and
calculating a weighted sum of the shifted, exposure-normalized images.

6. The at least one non-transitory machine-readable medium of claim 5 , weightings of the shifted, exposure-normalized images determined using a midpoint of an intensity scale for all pixel locations of the image.

7. The at least one non-transitory machine-readable medium of claim 5 , weightings of the shifted, exposure-normalized images proportional to exposure levels of a respective image of the array of images.

8. The at least one non-transitory machine-readable medium of claim 1 , an image of the array of images comprising raw Bayer data from a respective camera.

9. The at least one non-transitory machine-readable medium of claim 1 , an image of the array of images comprising output data of an imaging pipeline from a respective camera.

10. The at least one non-transitory machine-readable medium of claim 1 , estimating the HDR image patch comprising usage of a camera response curve.

11. An apparatus to produce a disparity estimate, comprising:
a processor circuit; and
an imaging management module for execution on the processor circuit to:
receive an array of images from a camera array, the array of images to comprise a plurality of images including at least a first image taken with a first value of an exposure parameter and a second image taken with a second value of the exposure parameter, the first value different than the second value;
estimate a noise level of each image of the array of images;
generate a plurality of exposure-normalized images, each one of the plurality of exposure-normalized images to be generated by normalizing a respective one of the plurality images based on an exposure parameter value used to capture that one of the plurality of images and a median exposure level for the plurality of images;
identify candidate disparities in each of the plurality of exposure-normalized images;
estimate a high dynamic range (HDR) image patch for each candidate disparity; and
compute an error from the HDR image patch and an objective function, to produce a disparity estimate.

12. The apparatus of claim 11 , the imaging management module to:
produce a composite image having a higher dynamic range than any one of the array of images, by use of the disparity estimate.

13. The apparatus of claim 11 , capturing a first image from the camera array with a first value of ISO gain and a second image from the camera array with a second value of ISO gain.

14. The apparatus of claim 11 , the camera array comprising one of a single row of cameras and a single column of cameras.

15. The apparatus of claim 11 , the imaging management module to:
designate a reference view as one image of the array of images;
compute a respective disparity between the reference view and each other image of the array of images;
shift each exposure-normalized image by its respective disparity; and
calculate a weighted sum of the shifted, exposure-normalized images.

16. The apparatus of claim 15 , the imaging management module to determine weightings of the shifted, exposure-normalized images using a midpoint of an intensity scale for all pixel locations of the image.

17. The apparatus of claim 15 , the imaging management module to determine weightings of the shifted, exposure-normalized images proportional to exposure levels of a respective image of the array of images.

18. The apparatus of claim 11 , an image of the array of images comprising raw Bayer data from a respective camera.

19. The apparatus of claim 11 , an image of the array of images comprising output data of an imaging pipeline from a respective camera.

20. The apparatus of claim 11 , the imaging management module to estimate the HDR image patch by usage of a camera response curve.

21. A system to produce a disparity estimate, comprising:
a processor circuit;
a camera array comprising a plurality of cameras; and
an imaging management module for execution on the processor circuit to:
capture an array of images from the camera array, the array of images to comprise a plurality of images including at least a first image taken with a first value of an exposure parameter and a second image taken with a second value of the exposure parameter, the first value different than the second value;
estimate a noise level of each image of the array of images;
generate a plurality of exposure-normalized images, each one of the plurality of exposure-normalized images to be generated by normalizing a respective one of the plurality images based on an exposure parameter value used to capture that one of the plurality of images and a median exposure level for the plurality of images;
identify candidate disparities in each of the plurality of exposure-normalized images;
estimate a high dynamic range (HDR) image patch for each candidate disparity; and
compute an error from the HDR image patch and an objective function, to produce a disparity estimate.

22. The system of claim 21 , the imaging management module to:
produce a composite image having a higher dynamic range than any one of the array of images, by use of the disparity estimate.

23. The system of claim 21 , the array of images comprising a first image taken with a first value of ISO

Assignees

• Intel Corp

Inventors

• Park Sung Hee
• Seshadrinathan Kalpana

Classifications

• H04N13/271

  wherein the generated image signals comprise depth maps or disparity maps · CPC title

• G06T2200/04

  involving 3D image data · CPC title

• G06T2207/20208

  High dynamic range [HDR] image processing · CPC title

• G06T7/55 Primary

  from multiple images · CPC title

• G06T5/50

  using two or more images, e.g. averaging or subtraction · CPC title