Automatic contrast enhancement

• US 7,953,286 B2
• Filed: 08/07/2007
• Issued: 05/31/2011
• Est. Priority Date: 08/08/2006
• Status: Active Grant

• Current Assignee: STMicroelectronics Asia Pacific Pte Limited (STMicroelectronics NV)
• Original Assignee: STMicroelectronics Asia Pacific Pte Limited (STMicroelectronics NV)
• Inventors: Chiang, Patricia Wei Yin, Hui, Yau Wai
• Primary Examiner(s): PATEL, KANJIBHAI B
• Application Number:
  US11/890,606
• Publication Number:
  US 20080037897A1
• Time in Patent Office: 1,393 Days
• Field of Search: 382/168, 382/172, 382/254, 382263-270, 382273-275, 382/312, 358521-522, 358/532
• US Class Current: 382/254
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 5/40 using histogram techniques
• CPC Class Codes: G06T 5/92 based on global image properties
• CPC Class Codes: H04N 5/20 Circuitry for controlling amplitude response

Abstract

In a process for enhancing contrast of an image having pixels in different brightness intensities, a histogram in discrete bins is generated. Each bin represents a pixel population of at least one pixel brightness intensity. A peak and a peak region of the histogram is then identified, wherein the peak region is a range of discrete bins around the peak. An average pixel population within the peak region is computed, and the pixel populations of the discrete bins within the peak region that exceeds the average pixel population are distributed. A transfer curve for mapping onto the image is then generated. The process can be used in an image processor for enhancing contrast of an image having pixel. Still further, a display having a receiver and a screen can include the foregoing image processor.

Claims

1. A process for enhancing contrast of an image having pixels in different brightness intensities, the process comprising:

generating a histogram in discrete bins, each bin representing a pixel population of at least one pixel brightness intensity;

computing a smoothened line curve from the generated histogram;

computing an average pixel population of the histogram;

identifying a peak and a peak region of the histogram, wherein the peak region is a range of discrete bins around the peak bounded by intersection of the smoothened line curve with the computed average pixel population of the histogram;

computing an average pixel population of the peak region;

distributing the pixel populations of the discrete bins within the peak region that exceed the average pixel population of the peak region to produce a re-distributed histogram, wherein distributing is limited to the discrete bins within the peak region; and

generating a transfer curve for mapping onto the image from the re-distributed histogram.

2. The process according to claim 1, wherein the histogram has 2ⁿ discrete bins, where n is number of bits of the image.

3. The process according to claim 1, further comprising expanding the histogram to provide 2ⁿ discrete bins when the generated histogram has a number of discrete bins less than 2ⁿ, where n is number of bits of the image.

4. The process according to claim 1, wherein computing a smoothened line curve comprises filtering the histogram with a smoothing filter for avoiding discontinuities of pixel population between successive discrete bins.

5. The process according to claim 4, wherein the smoothing filter comprises smoothing factors, and wherein the smoothing filter provides relatively higher smoothing factors at mid level pixel intensities than that of high and low pixel intensities.

6. The process according to claim 5, wherein the smoothing filter further comprises a smearing parameter for de-emphasizing the histogram.

7. The process according to claim 6, wherein the peak, the peak region, and the peak density are identified based on the filtered histogram.

8. The process according to claim 1, wherein distributing comprises distributing the pixel populations which exceed the average pixel population of the peak region along two sides of the peak.

9. The process according to claim 1, wherein distributing further comprises truncating overflows of the distributed pixel population beyond the peak region.

10. The process according to claim 1, further comprising computing an adaptive gain based on the peak, the peak region and the average pixel population.

11. The process according to claim 10, wherein the adaptive gain is computed based on a peak density factor provided by formula

12. The process according to claim 10, wherein the adaptive gain is computed based on a peak position factor provided by formula

13. The process according to claim 10, wherein the adaptive gain is computed based on a peak distribution factor is provided by formula

14. The process according to claim 1, wherein the transfer curve comprises two opposing banded curves separated by a pivot point that corresponds to the peak.

15. The process according to claim 14, wherein the transfer curve is provided by formulas

16. The process according to claim 15, wherein the transfer curve is provided by formulas

17. The process according to claim 1, wherein the histogram is provided based on an active window of the image.

18. The process according to claim 1, wherein the average pixel population of the peak region is an average of a total pixel population over the number of discrete bins within the peak region.

19. The process according to claim 1, wherein the process is applied on a video comprising video frames of images.

20. The process according to claim 19, wherein computing the smoothened line curve comprises:

computing a sum of absolute difference (SAD) of the histogram of a current image based on a previous image; and

performing a SAD weight recursive temporal filtering to the histogram.

21. The process according to claim 19, further comprising:

computing a SAD of a transfer curve; and

performing a SAD weighted recursive temporal filtering on the transfer curve.

22. An image processor for enhancing contrast of an image having pixels comprising:

a histogram generator adapted to generate a histogram in discrete bins, each bin representing a pixel population of at least one pixel brightness intensity;

a histogram pre-processor adapted to compute a smoothened line curve for the generated histogram;

a peak analyzer adapted to compute an average pixel population of the histogram, identify a peak and a peak region, and compute an average pixel population of the peak region, wherein the peak region is a range of discrete bins around the peak bounded by intersection of the smoothened line curve with the computed average pixel population of the histogram;

a pixel distributor adapted to distribute the pixel populations of the discrete bins within the peak region that exceed the average pixel population of the peak region to produce a re-distributed histogram, wherein distributing is limited to the discrete bins within the peak region; and

a transfer curve generator adapted to generate from the re-distributed histogram a transfer curve for mapping onto the image.

23. The image processor according to claim 22, wherein the histogram has 2ⁿ discrete bins, where n is number of bits of the image.

24. The image processor according to claim 22, wherein the histogram is expanded to provide 2ⁿ discrete bins when the generated histogram has a number of discrete bins less than 2ⁿ, where n is number of bits of the image.

25. The image processor according to claim 22, wherein the histogram pre-processor computes the smoothened line curve using a smoothing filter adapted to filter the histogram for avoiding discontinuities of pixel population between successive discrete bins.

26. The image processor according to claim 25, wherein the smoothing filter comprises smoothing factors, and wherein the smoothing filter provides relatively higher smoothing factors at mid level pixel intensities than that of high and low pixel intensities.

27. The image processor according to claim 26, wherein the smoothing filter further comprises a smearing parameter for de-emphasizing the histogram.

28. The image processor according to claim 27, wherein the peak analyzer is adapted to identify the peak, the peak region, and the peak density based on the filtered histogram.

29. The image processor according to claim 22, wherein the pixel distributor is adapted to distribute the pixel populations which exceed the average pixel population of the peak region along two sides of the peak.

30. The image processor according to claim 22, wherein the pixel distributor is adapted to truncate overflows of the distributed pixel population beyond the peak region.

31. The image processor according to claim 22, further comprising an adaptive gain calculator adapted to compute an adaptive gain based on the peak, the peak region and the average pixel population.

32. The image processor according to claim 31, wherein the adaptive gain is computed based on a peak density factor provided by formula

33. The image processor according to claim 31, wherein the adaptive gain is computed based on a peak position factor provided by formula

34. The image processor according to claim 31, wherein the adaptive gain is computed based on a peak distribution factor is provided by formula

35. The image processor according to claim 22, wherein the transfer curve comprises two opposing banded curves separated by a pivot point that corresponds to the peak.

36. The image processor according to claim 35, wherein the transfer curve is provided by formulas

37. The image processor according to claim 36, wherein the transfer curve is provided by formulas

38. The image processor according to claim 22, wherein the histogram is provided based on an active window of the image.

39. The image processor according to claim 22, wherein the average pixel population of the peak region is an average of a total pixel population over the number of discrete bins within the peak region.

40. The image processor according to claim 22, wherein the image processor is use for processing a video comprising video frames of images.

41. The image processor according to claim 40, wherein the histogram preprocessor further comprises:

means for computing a sum of absolute difference (SAD) of the histogram of a current image based on a previous image; and

a SAD weight recursive temporal filter for performing a SAD weight recursive temporal filtering to the histogram.

42. The process according to claim 40, further comprising

means for computing a SAD of a transfer curve;

and

a SAD weight recursive temporal filter for performing a SAD weighted recursive temporal filtering on the transfer curve.

43. A process for enhancing contrast of an image having pixels in different brightness intensities, the process comprising:

generating a histogram in discrete bins, each bin representing a pixel population of at least one pixel brightness intensity;

computing an average pixel population of the histogram;

identifying a peak and a peak region of the histogram, wherein the peak region is a range of discrete bins around the peak defined between a lower bin and an upper bin wherein the pixel population for each of the discrete bins of the peak region exceeds the computed average pixel population of the histogram;

computing an average pixel population of the peak region;

distributing the pixel populations of the discrete bins within the peak region that exceed the average pixel population of the peak region to produce a re-distributed histogram, wherein distributing is limited solely to the discrete bins within the peak region and any excess pixel populations which cannot be distributed to the discrete bins within the peak region are truncated; and

generating a transfer curve for mapping onto the image from the re-distributed histogram.