Method of adaptive image contrast enhancement

• US 7,512,269 B2
• Filed: 08/04/2005
• Issued: 03/31/2009
• Est. Priority Date: 08/04/2005
• Status: Active Grant

• Current Assignee: Lone Star Technological Innovations, LLC (TGT Enterprises, LLC)
• Original Assignee: Oplus Technologies Ltd. (Intel Corporation)
• Inventors: Levy, Avraham, Golan, Asaf
• Primary Examiner(s): Couso, Jose L
• Application Number:
  US11/196,320
• Publication Number:
  US 20070031055A1
• Time in Patent Office: 1,335 Days
• Field of Search: 382166-172, 382/254, 382270-271, 382/274, 348671-674, 358 321- 327, 358520-522
• US Class Current: 382/168
• CPC Class Codes: G06T 5/40 using histogram techniques
• CPC Class Codes: G06T 5/92 based on global image properties

Abstract

The contrast of a digital image is enhanced by forming a histogram of the luma values of an active area of the image, computing “initial” bin indices that partition the histogram into bins of substantially equal population, mapping the “initial” bin indices into “final” bin indices that are more uniformly spaced than the “initial” bin indices, and adjusting the luma values of all the pixels in accordance with the “final” bin indices.

Claims

1. For an image that includes a plurality of pixels, each pixel having a respective initial luma value and two respective initial chroma values, a method of adjusting the initial luma values, comprising the steps of using a computer to implement the steps of:

(a) forming a histogram of the initial luma values of at least a portion of the pixels;

(b) computing a plurality of initial bin indices that partitions said histogram into a plurality of bins that have substantially equal populations;

(c) mapping said initial bin indices into a plurality of final bin indices that are spaced more uniformly than said initial bin indices; and

(d) adjusting the initial luma values in accordance with said final bin indices, thereby obtaining, for each pixel, a respective final luma value.

2. The method of claim 1, wherein said histogram is formed using initial luma values of only a portion of the pixels.

3. The method of claim 1, wherein said final bin indices are substantially equally spaced.

4. The method of claim 3, wherein the initial luma values are adjusted in accordance with said initial bin indices.

5. The method of claim 4, wherein each said initial luma value is adjusted by steps including:

(i) determining a pair of said initial bin indices that bracket said each initial luma value; and

(ii) transforming said each initial luma value into a respective final luma value that is bracketed by a pair of said final bin indices that corresponds to said pair of initial bin values.

6. The method of claim 5, wherein said transforming is piecewise linear.

7. The method of claim 1, wherein said final bin indices are determined a postiori.

8. The method of claim 7, wherein said mapping includes modifying said initial bin indices, thereby obtaining a corresponding plurality of modified bin indices that are spaced more uniformly than said initial bin indices, the initial luma values then being adjusted in accordance with said modified bin indices.

9. The method of claim 8, wherein each said initial luma value is adjusted by steps including:

(i) determining a pair of said modified bin indices that bracket said each initial luma value; and

(ii) transforming said each initial luma value into a respective final luma value that is bracketed by a pair of said final bin indices that corresponds to said pair of modified bin values.

10. The method of claim 9, wherein said transforming is piecewise linear.

11. The method of claim 1, wherein said mapping is effected by steps including:

(i) mapping said initial bin indices into a plurality of intermediate bin indices that are substantially equally spaced; and

(ii) mapping said intermediate bin indices into said final bin indices in accordance with a desired degree of contrast gain.

12. For an image that includes a plurality of pixels, each pixel having a respective initial value, a method of adjusting the initial values, comprising the steps of using a computer to implement the steps of:

(a) forming a histogram of the initial values of at least a portion of the pixels;

(b) computing a plurality of initial bin indices that partitions said histogram into a plurality of bins that have substantially equal populations;

(c) mapping said initial bin indices into a plurality of final bin indices that are spaced more uniformly than said initial bin indices; and

(d) adjusting the initial values in accordance with said final bin indices, thereby obtaining, for each pixel, a respective final value.

13. The method of claim 12, wherein said histogram is formed using initial values of only a portion of the pixels.

14. The method of claim 12, wherein said final bin indices are substantially equally spaced.

15. The method of claim 14, wherein the initial values are adjusted in accordance with said initial bin indices.

16. The method of claim 15, wherein each said initial value is adjusted by steps including:

(i) determining a pair of said initial bin indices that bracket said each initial value; and

(ii) transforming said each initial value into a respective final value that is bracketed by a pair of said final bin indices that corresponds to said pair of initial bin values.

17. The method of claim 16, wherein said transforming is piecewise linear.

18. The method of claim 12, wherein said final bin indices are determined a postiori.

19. The method of claim 18, wherein said mapping includes modifying said initial bin indices, thereby obtaining a corresponding plurality of modified bin indices that are spaced more uniformly than said initial bin indices, the initial values then being adjusted in accordance with said modified bin indices.

20. The method of claim 19, wherein each said initial value is adjusted by steps including:

(i) determining a pair of said modified bin indices that bracket said each initial value; and

(ii) transforming said initial value into a respective final value that is bracketed by a pair of said final bin indices that corresponds to said pair of modified bin values.

21. The method of claim 20, wherein said transforming is piecewise linear.

22. The method of claim 12, wherein said mapping is effected by steps including:

(i) mapping said initial bin indices into a plurality of intermediate bin indices that are substantially equally spaced; and

(ii) mapping said intermediate bin indices into said final bin indices in accordance with a desired degree of contrast gain.

23. A system, comprising:

(a) a memory for storing an image that includes a plurality of pixels, each pixel having a respective value; and

(b) a processor for;

(i) forming a histogram of said values of at least a portion of the pixels,

(ii) computing a plurality of initial bin indices that partitions said histogram into a plurality of bins that have substantially equal populations,

(iii) mapping said initial bin indices into a plurality of final bin indices that are spaced more uniformly than said initial bin indices, and

(iv) adjusting said values in accordance with said final bin indices.

24. The system of claim 23, wherein said image is a monochrome image.

25. The system of claim 23, wherein said image is a color image, said values being initial luma values.

26. A computer readable storage medium having computer readable code embodied on said computer readable storage medium, the computer readable code for enabling a computer to process an image that includes a plurality of pixels, each pixel having a respective value, the computer readable code comprising:

(a) program code for forming a histogram of the values of at least a portion of the pixels;

(b) program code for computing a plurality of initial bin indices that partitions said histogram into a plurality of bins that have substantially equal populations;

(c) program code for mapping said initial bin indices into a plurality of final bin indices that are spaced more uniformly than said initial bin indices; and

(d) program code for adjusting said values in accordance with said final bin indices.