Method for estimating and adjusting digital image contrast

• US 5,822,453 A
• Filed: 12/10/1996
• Issued: 10/13/1998
• Est. Priority Date: 12/10/1996
• Status: Expired due to Fees

• Current Assignee: Eastman Kodak Company
• Original Assignee: Eastman Kodak Company
• Inventors: Kwon, Heemin, Lee, Hsien-Che
• Primary Examiner(s): Mancuso, Joseph
• Assistant Examiner(s): AHMED, SAMIR ANWAR
• Application Number:
  US08/763,270
• Time in Patent Office: 672 Days
• Field of Search: 382/167, 382/168, 382/169, 382/224, 382/274, 382/279, 382/170, 358/406, 358/448, 358/455, 358/467, 358/515, 358/519, 358/521, 358/524
• US Class Current: 382/169
• CPC Class Codes: G06T 5/92 based on global image properties
• CPC Class Codes: H04N 1/4074 using histograms

Abstract

A method of estimating the scene contrast from a digital image, the method comprises the steps of: forming a Laplacian histogram distribution; determining, from the Laplacian histogram, first and second thresholds which eliminate substantially uniform areas or a substantially textured portion of the digital image; selecting pixels which are based on the first and second thresholds from the digital image; forming a histogram from the sampled pixels; computing a standard deviation of the sampled histogram; and estimating contrast of the digital image by comparing the computed standard deviation with a predetermined contrast for determining contrast of the input image in relationship with the predetermined contrast.

Claims

1. A method of estimating the scene contrast from a digital image, the method comprising the steps of:

a) determining a Laplacian value for value for each pixel of the digital image;

b) forming a Laplacian histogram distribution based on the Laplacian values;

c) determining, from the Laplacian histogram, first and second thresholds which eliminate substantially uniform areas or substantially textured portion of the digital image;

d) selecting pixels which are based on the first and second thresholds from the digital image;

e) forming an intensity histogram from the selected pixels;

f) computing a standard deviation of the intensity histogram; and

g) estimating contrast of the digital image by comparing the computed standard deviation with a predetermined contrast for determining contrast of the input image in relationship with the predetermined contrast.

2. The method of claim 1, wherein the step of determining the first and second thresholds includes finding a local maxima of a second derivative of the Laplacian histogram.

3. The method of claim 1, wherein the step of selecting the pixels includes selecting the pixels that exceed substantially the first threshold.

4. The method of claim 1, wherein the step of selecting the pixels includes selecting the pixels that are less than substantially the second threshold.

5. The method of claim 1 further comprising the step of computing a tone scale transformation curve from the estimate contrast of the digital image.

6. The method of claim 5 further comprising the step of constructing the tone transformation curve by modifying the intensity histogram formed from the selected pixels so that a target histogram is generated which said target histogram is dependent on the digital image.

7. The method of claim 6 further comprising the step of constructing a tone transformation curve by convolving the intensity histogram with another pre-defined function with image-dependent parameters.

8. The method of claim 7, wherein the predefined function is a Gaussian distribution.

9. The method of claim 1 further comprising forming a target histogram from the intensity histogram by convolving the intensity histogram with a pre-defined function, and constructing a tone scale curve by transforming the intensity histogram to the target histogram.

10. The method of claim 9, wherein the step of determining the first and second thresholds includes finding a local maxima of a second derivative of the Laplacian histogram.

11. The method of claim 9, wherein the step of selecting the pixels includes selecting the pixels that exceed substantially the first threshold.

12. The method of claim 9, wherein the step of selecting the pixels includes selecting the pixels that are less than substantially the second threshold.

13. A method of estimating the scene contrast from a digital image, the method comprising the steps of:

a) determining a Laplacian value for each pixel of the digital image;

b) forming a Laplacian histogram distribution based on the Laplacian values;

c) determining, from the Laplacian histogram, first and second thresholds;

d) selecting pixels which are based on the first and second thresholds from the digital image;

e) forming an intensity histogram from the selected pixels;

f) forming a target histogram from the intensity histogram by convolving the intensity histogram with a pre-defined function; and

g) constructing a tone scale curve by transforming the intensity histogram to the target histogram.

14. The method as in claim 13, wherein the step of forming a target histogram includes convolving the intensity histogram with a Gaussian distribution.

15. The method as in claim 14, wherein the step of forming a target histogram includes convolving the intensity histogram with the Gaussian distribution having a standard deviation substantially the same as the sampled histogram.

16. The method as in claim 13 further comprising the step of computing a standard deviation of the intensity histogram and estimating contrast of the digital image by comparing the computed standard deviation with a predetermined contrast for determining contrast of the input image in relationship with the predetermined contrast.

17. The method of claim 16, wherein the step of determining the first and second thresholds includes finding a local maxima of a second derivative of the Laplacian histogram.

18. The method of claim 17, wherein the step of selecting the pixels includes selecting the pixels that exceed substantially the first threshold.

19. The method of claim 18, wherein the step of selecting the pixels includes selecting the pixels that are less than substantially the second threshold.

20. The method of claim 19 further comprising the step of constructing the tone transformation curve by modifying the intensity histogram formed from the selected pixels so that an image-dependent target histogram is generated.

21. A computer program product for excluding pixels from a uniform portion or a texture portion of a digital image, comprising a computer readable storage medium having a computer program stored thereon for performing the steps of:

a) determining a Laplacian value for each pixel of the digital image;

b) forming a Laplacian histogram distribution based on the Laplacian values;

c) determining, from the Laplacian histogram, first and second thresholds which eliminate substantially the uniform portion or substantially the textured portion of the digital image; and

d) selecting pixels which are based on the first and second thresholds from the digital image.

22. A method for excluding pixels from a uniform portion or a texture portion of a digital image, comprising the steps of:

a) determining a Laplacian value for each pixel of the digital image;

b) forming a Laplacian histogram distribution based on the Laplacian values;

c) determining, from the Laplacian histogram, first and second thresholds which eliminate substantially the uniform portion or substantially the textured portion of the digital image; and

d) selecting pixels which are based on the first and second thresholds from the digital image.

23. The method of claim 22, wherein the step of determining the first and second thresholds includes finding a local maxima of a second derivative of the Laplacian histogram.

24. A computer program product for estimating the scene contrast from a digital image, comprising:

a computer readable storage medium having a computer program stored thereon for performing the steps of;

a) determining a Laplacian value for each pixel of the digital image;

b) forming a Laplacian histogram distribution based on the Laplacian values;

c) determining, from the Laplacian histogram, first and second thresholds which eliminate substantially uniform areas or substantially textured portion of the digital image;

d) selecting pixels which are based on the first and second thresholds from the digital image;

e) forming an intensity histogram from the selected pixels;

f) computing a standard deviation of the intensity histogram; and

g) estimating contrast of the digital image by comparing the computed standard deviation with a predetermined contrast for determining contrast of the input image in relationship with the predetermined contrast.

25. The computer program product as in claim 24, wherein the step of determining the first and second thresholds includes finding a local maxima of a second derivative of the Laplacian histogram.

26. The computer program product as in claim 24, wherein the step of selecting the pixels includes selecting the pixels that exceed substantially the first threshold.

27. The computer program product as in claim 24, wherein the step of selecting the pixels includes selecting the pixels that are less than substantially the second threshold.

28. The computer program product as in claim 24 further comprising the step of computing a tone scale transformation curve from the estimate contrast of the digital image.

29. The computer program product as in claim 28 further comprising the step of constructing the tone transformation curve by modifying the intensity histogram formed from the selected pixels so that a target histogram is generated which said target histogram is dependent on the digital image.

30. The computer program product as in claim 29 further comprising the step of constructing a tone transformation curve by convolving the intensity histogram with another pre-defined function with image-dependent parameters.

31. The computer program product as in claim 30, wherein the pre-defined function is a Gaussian distribution.

32. The computer program product as in claim 24 further comprising forming a target histogram from the intensity histogram by convolving the intensity histogram with a pre-defined function, and constructing a tone scale curve by transforming the intensity histogram to the target histogram.

33. The computer program product as in claim 32, wherein the step of determining the first and second thresholds includes finding a local maxima of a second derivative of the Laplacian histogram.

34. The computer program product as in claim 32, wherein the step of selecting the pixels includes selecting the pixels that exceed substantially the first threshold.

35. The computer program product as in claim 32 wherein the step of selecting the pixels includes selecting the pixels that are less than substantially the second threshold.

36. A computer program product for estimating the scene contrast from a digital image, comprising:

a computer readable storage medium having a computer program stored thereon for performing the steps of;

a) determining a Laplacian value for each pixel of the digital image;

b) forming a Laplacian histogram distribution based on the Laplacian values;

c) determining, from the Laplacian histogram, first and second thresholds;

d) selecting pixels which are based on the first and second thresholds from the digital image;

e) forming an intensity histogram from the selected pixels;

f) forming a target histogram from the intensity histogram by convolving the intensity histogram with a predefined function; and

g) constructing a tone scale curve by transforming the intensity histogram to the target histogram.

37. The computer program product as in claim 36, wherein the step of forming a target histogram includes convolving the intensity histogram with a Gaussian distribution.

38. The computer program product as in claim 37, wherein the step of forming a target histogram includes convolving the intensity histogram with the Gaussian distribution having a standard deviation substantially the same as the sampled histogram.

39. The computer program product as in claim 36 further comprising the step of computing a standard deviation of the intensity histogram and estimating contrast of the digital image by comparing the computed standard deviation with a predetermined contrast for determining contrast of the input image in relationship with the predetermined contrast.

40. The computer program product as in claim 39, wherein the step of determining the first and second thresholds includes finding a local maxima of a second derivative of the Laplacian histogram.

41. The computer program product as in claim 40, wherein the step of selecting the pixels includes selecting the pixels that exceed substantially the first threshold.

42. The computer program product as in claim 41, wherein the step of selecting the pixels includes selecting the pixels that are less than substantially the second threshold.

43. The computer program product as in claim 42 further comprising the step of constructing the tone transformation curve by modifying the intensity histogram formed from the selected pixels so that an image-dependent target histogram is generated.