Method for efficient rate control

• US 6,356,668 B1
• Filed: 12/29/1998
• Issued: 03/12/2002
• Est. Priority Date: 12/29/1998
• Status: Expired due to Term

• Current Assignee: Monument Peak Ventures, LLC (Dominion Harbor Enterprises, LLC)
• Original Assignee: Eastman Kodak Company
• Inventors: Rabbani, Majid, Honsinger, Chris W., Jones, Paul W.
• Primary Examiner(s): Tran, Phuoc
• Assistant Examiner(s): Alavi, Amir
• Application Number:
  US09/222,483
• Time in Patent Office: 1,169 Days
• Field of Search: 382/232, 382/233, 382/234, 382/240, 382/241, 382/242, 382/243, 382/244, 382/245, 382/246, 382/247, 382/248, 382/250, 382/251, 382/239, 340/825.8, 348/405, 348/419, 348/420, 358/432, 358/433, 375/240.11, 375/240.2, 375/240.24
• US Class Current: 382/251
• CPC Class Codes: G06T 9/00 Image coding bandwidth or redundancy reduction for static pictures H04N1/41; coding or decoding of static colour picture signals H04N1/64; methods or arrangements for coding, decoding, compressing or decompressing digital video signals H04N19/00

Abstract

An efficient method for determining the compressed rate of a digital image with a given set of compression parameters, the method comprises the steps of forming a plurality of non-overlapping image regions, each of which includes a plurality of contiguous pixels, from the digital image where the total number of pixels contained in all of the contiguous regions is less than the total pixels in the digital image; compressing each region with the given set of compression parameters; and, combining the resulting rate from each region to form an estimate of the compressed rate of the digital image.

Claims

1. An efficient method for determining the compressed rate of a digital image with a given set of compression parameters that affect the amount of quantization, comprising the steps of:

a) forming a plurality of sparsely-sampled non-overlapping image regions, each of which includes a plurality of contiguous pixels, from the digital image where the total number of pixels contained in all of the regions is less than the total pixels in the digital image;

b) compressing each region with the given set of compression parameters, thereby generating a resulting rate for each region; and

,

c) combining the resulting rate from each region to form an estimate of the compressed rate of the digital image.

2. The method as in claim 1, wherein step (a) includes using image regions all of which are the same shape and same size.

3. The method as in claim 2, wherein the step of using image regions all of which are the same shape and same size includes using rectangular-shaped blocks.

4. The method as in claim 3, wherein the step of using the rectangular-shaped blocks includes using 8×

8 blocks.

5. The method as in claim 1 further comprising the step of placing the image regions at a plurality of uniformly spaced intervals in the digital image.

6. The method as in claim 2 further comprising the step of placing the same shape and same size image regions at a plurality of uniformly spaced intervals in the digital image.

7. The method as in claim 3 further comprising the step of placing the rectangular-shaped blocks at a plurality of uniformly spaced intervals in the digital image.

8. The method as in claim 4 further comprising the step of placing the 8×

8 blocks at a plurality of uniformly spaced intervals in the digital image.

9. The method as in claim 1 further comprising the step of placing the image regions at a plurality of randomly selected positions in the digital image.

10. The method as in claim 2 further comprising the step of placing the same shape and same size image regions at a plurality of randomly selected positions in the digital image.

11. The method as in claim 3 further comprising the step of placing the rectangular-shaped blocks at a plurality of randomly selected positions in the digital image.

12. The method as in claim 4 further comprising the step of placing the 8×

8 blocks at a plurality of randomly selected positions in the digital image.

13. The method as in claim 1 further comprising the steps of dividing the digital image into a plurality of equally sized, contiguous and non-overlapping tiles, and placing one image region in each of the tiles.

14. The method as in claim 2 further comprising the steps of dividing the digital image into a plurality of equally sized, contiguous and non-overlapping tiles, and placing the image regions all of which are the same shape and same size, in each of the tiles.

15. The method as in claim 3 further comprising the steps of dividing the digital image into a plurality of equally sized, contiguous and non-overlapping tiles, and placing the rectangular-shaped blocks in each of the tiles.

16. The method as in claim 4 further comprising the steps of dividing the digital image into a plurality of equally sized, contiguous and non-overlapping tiles, and placing the 8×

8 blocks in each of the tiles.

17. The method as in claim 14, further comprising the steps of placing the image regions in each of the tiles in a random fashion.

18. The method as in claim 15, further comprising the steps of placing the rectangular shaped blocks in each of the tiles in a random fashion.

19. The method as in claim 16, further comprising the steps of placing the 8×

8 blocks in each of the tiles in a random fashion.

20. The method as in claim 4, where the step of compressing the 8×

8 image block includes using JPEG compression.

21. The method as in claim 8, where the step of compressing the 8×

8 image block includes using JPEG compression.

22. The method as in claim 12, where the step of compressing the 8×

8 image block includes using JPEG compression.

23. The method as in claim 16, where the step of compressing the 8×

8 image block includes using JPEG compression.

24. The method as in claim 19, where the step of compressing the 8×

8 image block includes using JPEG compression.

25. The method as in claim 20, further comprising the step of using a specification of a JPEG Q-table as the compression parameter.

26. The method as in claim 21, further comprising the step of using a specification of a JPEG Q-table as the compression parameter.

27. The method as in claim 22, further comprising the step of using a specification of a JPEG Q-table as the compression parameter.

28. The method as in claim 23, further comprising the step of using a specification of a JPEG Q-table as the compression parameter.

29. The method as in claim 24, further comprising the step of using a specification of a JPEG Q-table as the compression parameter.

30. The method as in claim 25, further comprising the step of repeating steps b) and c) for a plurality of Q-table specifications to build a rate-distortion characteristic curve.

31. The method as in claim 26, further comprising the step of repeating steps b) and c) for a plurality of Q-table specifications to build a rate-distortion characteristic curve.

32. The method as in claim 27, further comprising the step of repeating steps b) and c) for a plurality of Q-table specifications to build a rate-distortion characteristic curve.

33. The method as in claim 28, further comprising the step of repeating steps b) and c) for a plurality of Q-table specifications to build a rate-distortion characteristic curve.

34. The method as in claim 29, further comprising the step of repeating steps b) and c) for a plurality of Q-table specifications to build a rate-distortion characteristic curve.

35. The method as in claim 30, wherein a distortion metric used in the rate-distortion curve is computed for each set of Q-table specifications.

36. The method as in claim 31, wherein a distortion metric used in the rate-distortion curve is computed for each set of Q-table specifications.

37. The method as in claim 32, wherein a distortion metric used in the rate-distortion curve is computed for each set of Q-table specifications.

38. The method as in claim 33, wherein a distortion metric used in the rate-distortion curve is computed for each set of Q-table specifications.

39. The method as in claim 34, wherein a distortion metric used in the rate-distortion curve is computed for each set of Q-table specifications.

40. An efficient method for determining the compressed rate of a digital image with a given set of compression parameters that affect the amount of quantization, comprising the steps of

a) forming a plurality of non-overlapping image regions, each of which includes a plurality of contiguous pixels, by sampling the digital image uniformly over a large scale and randomly over a smaller scale;

b) compressing each region with the given set of compression parameters, thereby generating a resulting rate for each region; and

c) combining the resulting rate from each region to form an estimate of the compressed rate of the digital image.

41. The method as in claim 40, wherein step (a) includes using image regions all of which are the same shape and same size.

42. The method as in claim 40, where step (b) includes using JPEG compression.

43. The method as in claim 40, further comprising the step of repeating steps b) and c) for a plurality of sets of compression parameters to build a rate-distortion characteristic curve.

44. An efficient method for determining the compressed rate of a digital image given JPEG compression with a given quantization table, comprising the steps of:

a) forming a plurality of non-overlapping image regions, each of which includes a plurality of contiguous pixels, by dividing a digital image into uniformly distributed contiguous non-overlapping tiles and then selecting an image region from within each tile, where the total number of pixels contained in a region is less than the total number of pixels in a tile containing that region;

b) compressing each region with the given quantization table; and

,

c) combining the resulting rate from each region to form an estimate of the compressed rate of the digital image.

45. The method as in claim 44, wherein step (a) includes using image regions all of which are the same shape and same size.

46. The method as in claim 44, wherein step (a) includes placing each image region at a randomly selected position within each tile.

47. The method as in claim 44, further comprising the step of repeating steps b) and c) for a plurality of quantization tables to build a rate—

distortion characteristic curve.