Foreground/Background Segmentation in Digital Images with Differential Exposure Calculations

• US 20100328486A1
• Filed: 09/13/2010
• Published: 12/30/2010
• Est. Priority Date: 08/16/2004
• Status: Active Grant

• Current Assignee: FotoNation Limited (Adeia Inc.)
• Original Assignee: Tessera Technologies Ireland Limited (Adeia Inc.)
• Inventors: Steinberg, Eran, Prilutsky, Yury, Bigioi, Petronel, Corcoran, Peter
• Granted Patent: US 7,912,285 B2
• Time in Patent Office: Days
• US Class Current: 348/222.1
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/10152 Varying illumination
• CPC Class Codes: G06T 2207/20224 Image subtraction
• CPC Class Codes: G06T 2207/30216 Redeye defect
• CPC Class Codes: G06T 5/77 Retouching; Inpainting; Scratch removal
• CPC Class Codes: G06T 7/0002 Inspection of images, e.g. flaw detection
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/174 involving the use of two or more images
• CPC Class Codes: G06T 7/194 involving foreground-background segmentation
• CPC Class Codes: G06T 7/40 Analysis of texture depth or shape recovery from texture G06T7/529
• CPC Class Codes: G06V 10/28 Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
• CPC Class Codes: G06V 20/10 Terrestrial scenes scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59
• CPC Class Codes: G06V 40/165 using facial parts and geometric relationships
• CPC Class Codes: G06V 40/193 Preprocessing; Feature extraction
• CPC Class Codes: H04N 1/624 Red-eye correction control of camera based on recognised facial parts H04N23/611
• CPC Class Codes: H04N 23/74 by influencing the scene brightness using illuminating means
• CPC Class Codes: H04N 23/80 Camera processing pipelines; Components thereof

Abstract

A digital segmentation method and apparatus determines foreground and/or background within at least one portion of a captured image. The determining includes comparing a captured image to a pre-captured or post captured reference image of nominally the same scene. One of the images is taken with flash and the other without. The system can be implemented as part of a digital camera acquisition chain having effective computation complexity.

Claims

1. (canceled)

2. A digital image acquisition system having no photographic film, comprising:

(a) an apparatus for capturing digital images, including a lens, an image sensor and a processor;

(b) a flash unit for providing illumination during image capture;

(c) an analysis tool for comparing at least a portion of a captured image and a reference image of nominally the same scene, at least one of said captured and reference images being taken with flash, said analysis tool providing a measure of relative differences in illumination between regions of said captured image and said reference image of groups of pixels within said scene; and

(d) a classification tool for segmenting a foreground region from a background region within said scene based on said measure.

3. A system according to claim 2, further comprising an exposure equalizer for substantially equalizing an overall level of exposure of at least said portion of said captured and reference images prior to analysis by said analysis tool.

4. A system according to claim 2, further comprising a segmentation tool for determining one or more regions that are indicative of the foreground region, or of the background region, or of the background region and the foreground region, within at least one portion of a captured image, and wherein said analysis tool is arranged to analyse said foreground region or said background region, or both.

5. A system according to claim 2, wherein said classification tool is responsive to said measure exceeding a high threshold to classify a region as a foreground region, and responsive to said measure not exceeding a low threshold to classify a region as a background region.

6. A system according to claim 5, wherein said high and low threshold are coincident.

7. A system according to claim 2, further comprising a segmentation tool for determining one or more regions that are indicative of the foreground region, or of the background region, or of the background region and the foreground region, within at least one portion of the captured image, wherein said determining comprises comparing said captured image and the same reference image or a different reference image of nominally the same scene, or both.

8. A system according to claim 7, wherein the captured and reference images have different pixel resolutions, and wherein the system further comprises a pixel matching tool which is operative prior to application of the segmentation tool for matching the pixel resolutions of the captured and reference images at least in respect of said at least one portion.

9. A system according to claim 8, wherein said pixel matching tool utilizes up-sampling of the image of lower resolution or sub-sampling of the image of higher resolution, or both.

10. A system according to claim 7, further comprising an alignment tool which is operative prior to application of the segmentation tool for aligning said regions of said captured and reference images at least in respect of said at least one portion.

11. A system according to claim 7, further comprising an exposure equalizer for substantially equalizing an overall level of exposure of said regions or all of said captured and reference images at least in respect of said at least one portion.

12. A system according to claim 11, wherein the substantially equalising of the overall level of exposure of said regions or all of said captured and reference image comprises simulating an ambient exposure of the captured image on the reference image.

13. A system according to claim 12, wherein the simulating of the ambient exposure of the captured image on the reference image comprises digitally simulating a one or a combination of aperture, acquisition speed, color transformations and gain of the captured image on the reference image.

14. A system according to claim 12, wherein the simulating of the ambient exposure of the captured image comprises individual, non-uniform manipulating of individual regions or color channels or combinations thereof.

15. A system according to claim 11, wherein the substantially equalising of the overall level of exposure of said captured and reference image comprises setting an ambient exposure of the reference image to match a calculated exposure of the captured image.

16. A system according to claim 11, wherein at least in respect of said at least one portion, the segmentation tool determines corresponding pixels in the captured and reference images whose values differ by less than a predetermined threshold, and designates segments of the captured image bounded by said determined pixels as foreground or background by comparing pixel values in a segment with pixel values in a corresponding segment of the reference image.

17. A system according to claim 11, wherein at least in respect of said at least one portion, the segmentation tool determines upper and lower thresholds based on a comparison of the overall level of exposure of the captured and reference images and designates pixels of the captured image as foreground or background according to whether their values are greater than the upper threshold or less than the lower threshold.

18. A system according to claim 11, wherein at least in respect of said at least one portion, the segmentation tool designates one or more segments of the captured image as foreground or background by comparing pixel values in the captured and reference images.

19. A system according to claim 11, wherein the reference image comprises a preview image having a lower pixel resolution than the captured image, and the captured image comprises an image taken with flash.

20. A system according to claim 7, further comprising a face detection module.

21. A system according to claim 7, further comprising a red-eye detection filter or a red eye correction filter or both, for selective application to the foreground region.

22. A system according to claim 21, further comprising a probability module for changing a probability of a redeye candidate region being an actual redeye region according to whether the candidate appears in the foreground or background of the captured image.

23. A system according to claim 2, further comprising a depth of focus module for reducing a perceived depth of focus according to whether a candidate region appears in the foreground or background of the captured image.

24. A system according to claim 2, wherein the reference image comprises a preview image.

25. A system according to claim 2, wherein the reference image comprises an image captured chronologically after said captured image.

26. A system according to claim 2, wherein the reference image comprises a combination of multiple reference-images.

27. A system according to claim 2, wherein said digital image acquisition system comprises a digital camera.

28. A system according to claim 2, wherein said digital image acquisition system comprises a combination of a digital camera and an external processing device.

29. A system as claimed in claim 28, wherein said segmentation tool is located within said external processing device.

30. A system according to claim 2, wherein said digital image acquisition system comprises a batch processing system including a digital printing device.

31. A system according to claim 2, wherein said digital image acquisition system comprises a batch processing system including a server computer.

32. A computer readable medium having code embodied therein for programming a processor to perform a method of analyzing a captured image, the computer readable medium comprising an analysis and classification tool for comparing at least a portion of a captured image and a reference image of nominally the same scene, at least one of said captured and reference images being taken with flash, said tool providing a measure of relative differences in illumination between regions of said captured image and said reference image, and said tool for segmenting a foreground region from a background region within said scene based on said measure.

33. The analysis and classification tool of claim 32, further comprising a digital segmentation tool for determining one or more regions that are indicative of the foreground region, or of the background region, or of the background region and the foreground region, within at least one portion of a captured image, wherein said determining comprises comparing said captured image and a reference image of nominally the same scene, and wherein at least one of said captured and reference images being taken with flash.

34. The analysis and classification tool of claim 33, wherein the captured and reference images have different pixel resolutions, and wherein the segmentation tool operates in conjunction with a pixel matching tool which is operative prior to application of the segmentation tool for matching the pixel resolutions of the captured and reference images at least in respect of said at least one portion.

35. The analysis and classification tool of claim 34, wherein said pixel matching tool utilizes up-sampling of the image of lower resolution or sub-sampling of the image of higher resolution, or both.

36. The analysis and classification tool of claim 33, wherein the segmentation tool operates in conjunction with an alignment tool which is operative prior to application of the segmentation tool for aligning said captured and reference images at least in respect of said at least one portion.

37. The analysis and classification tool of claim 33, wherein said segmentation tool operates in conjunction with an exposure equalizer for substantially equalizing an overall level of exposure of said captured and reference images at least in respect of said at least one portion.

38. The analysis and classification tool of claim 33, wherein said segmentation tool operates in conjunction with a face detection module or a red-eye filter, or both, for selective application to the foreground region.

39. The analysis and classification tool of claim 38, wherein said segmentation tool further operates in conjunction with a probability module for changing a probability of a redeye candidate region being an actual redeye region according to whether the candidate appears in the foreground or background of the captured image.

40. The analysis and classification tool of claim 38, wherein said segmentation tool further operates in conjunction with a depth of focus module for reducing a perceived depth of focus according to whether a candidate region appears in the foreground or background of the captured image.

41. The analysis and classification tool of claim 38, wherein said segmentation tool further operates in conjunction with a blurring module for blurring said regions indicative of background of the captured image.

42. The analysis and classification tool of claim 32, wherein the reference image comprises a preview image.

43. The analysis and classification tool of claim 32, wherein the reference image comprises an image captured chronologically after said captured image.

44. The analysis and classification tool of claim 32, wherein the reference image comprises a combination of multiple reference-images.

45. A method of analyzing a captured image, comprising:

comparing at least a portion of a captured image and a reference image of nominally the same scene, at least one of said captured and reference images being taken with flash,

providing a measure of relative differences in illumination between regions of said captured image and said reference image, and

segmenting a foreground region from a background region within said scene based on said measure.

46. The method of claim 45, further comprising determining one or more regions that are indicative of the foreground region, or of the background region, or of the background region and the foreground region, within at least one portion of a captured image, wherein said determining comprises comparing said captured image and a reference image of nominally the same scene, and wherein at least one of said captured and reference images being taken with flash.

47. The method of claim 46, wherein the captured and reference images have different pixel resolutions, and wherein the method further comprises matching the pixel resolutions of the captured and reference images at least in respect of said at least one portion.

48. The method of claim 47, wherein said matching of pixel resolutions comprises up-sampling of the image of lower resolution or sub-sampling of the image of higher resolution, or both.

49. The method of claim 46, further comprising aligning said captured and reference images at least in respect of said at least one portion.

50. The method of claim 46, further comprising approximately equalizing an overall level of exposure of said captured and reference images at least in respect of said at least one portion.

51. The method of claim 46, further comprising detecting a face or filtering red eye, or both, within the foreground region.

52. The method of claim 51, further comprising changing a probability of a redeye candidate region being an actual redeye region according to whether the candidate appears in the foreground or background of the captured image.

53. The method of claim 51, further comprising reducing a perceived depth of focus according to whether a candidate region appears in the foreground or background of the captured image.

54. The method of claim 51, further comprising blurring said regions indicative of background of the captured image.

55. The method of claim 45, wherein the reference image comprises a preview image.

56. The method of claim 45, wherein the reference image comprises an image captured chronologically after said captured image.

57. The method of claim 45, wherein the reference image comprises a combination of multiple reference-images.