Methods and systems for complexity-based segmentation refinement

• US 20060002623A1
• Filed: 11/19/2004
• Published: 01/05/2006
• Est. Priority Date: 06/30/2004
• Status: Active Grant

• Current Assignee: Sharp Kabushiki Kaisha (Hon Hai Precision Industry Co., Ltd.)
• Original Assignee: Sharp Laboratories of America Incorporated (Hon Hai Precision Industry Co., Ltd.)
• Inventors: Matsuda, Toyohisa, Campbell, Richard John, Ferman, Ahmet Mufit, Dolan, John Edward
• Granted Patent: US 7,266,243 B2
• Time in Patent Office: Days
• US Class Current: 382/254
• CPC Class Codes: G06V 30/413 Classification of content, e.g. text, photographs or tables

Abstract

Embodiments of the present invention comprise methods and systems for image complexity estimation and selective complexity-based image processing.

Claims

1. A method for processing of a digital image, said method comprising:

a) performing an initial segmentation process on a non-object-based image data set;

b) estimating a complexity measure for said image data set based on the results of said initial segmentation process; and

c) performing a refined segmentation process on said image data set when said complexity measure meets a criterion.

2. A method as described in claim 1 further comprising performing a segmentation-based image enhancement process using a result of said refined segmentation process when said complexity measure meets said criterion.

3. A method as described in claim 1 wherein global image enhancement is performed when said complexity measure does not meet said criterion.

4. A method as described in claim 3 wherein said criterion represents a condition at which segmentation-based enhancement provides no discernible advantage over global enhancement.

5. A method as described in claim 1 wherein said criterion comprises said complexity remaining below a threshold value.

6. A method for processing of a digital image, said method comprising:

a) performing an initial segmentation process on an image;

b) obtaining process performance data for a refined segmentation process wherein said performance data relates to image complexity;

c) estimating a complexity measure for said image based on the results of said initial segmentation process and said performance data; and

d) performing said refined segmentation process when said complexity measure meets a criterion.

7. A method as described in claim 6 further comprising performing segmentation-based image enhancement using a result of said refined segmentation process when said complexity measure meets said criterion.

8. A method as described in claim 6 further comprising performing global image enhancement when said complexity measure does not meet said criterion.

9. A method as described in claim 6 further comprising performing global image enhancement when said complexity measure meets an alternative criterion.

10. A method as described in claim 6 wherein said criterion represents a condition at which segmentation-based enhancement provides no discernible advantage over global enhancement.

11. A method as described in claim 6 wherein said criterion comprises said complexity remaining below a threshold value.

12. A method for processing of a digital image, said method comprising:

a) performing an initial segmentation on a non-object-based image data set;

b) obtaining process performance data for a refined segmentation process wherein said performance data relates to image complexity;

c) estimating a complexity measure for said image data set based on the results of said initial segmentation and said performance data;

d) performing said refined segmentation process and performing segmentation-based image enhancement using a result of said refined segmentation process when said complexity measure meets a criterion.

13. A method as described in claim 12 wherein global image enhancement is performed when said complexity measure does not meet said criterion.

14. A method as described in claim 12 wherein said criterion represents a condition at which segmentation-based enhancement provides no discernible advantage over global enhancement.

15. A method as described in claim 12 wherein said criterion comprises said complexity remaining below a threshold value.

16. An apparatus for processing a digital image, said apparatus comprising:

a) an initial segmentation processor for performing an initial segmentation process on non-object-based data representing an image;

b) an estimator for estimating a complexity measure for said image based on the results of said initial segmentation; and

c) a refined segmentation processor for performing a refined segmentation process when said complexity measure meets a criterion.

17. A method as described in claim 16 further comprising a segmentation-based enhancement processor for performing segmentation-based image enhancement based on a result of said refined segmentation process when said complexity measure meets said criterion.

18. A method as described in claim 16 wherein said criterion comprises said complexity measure remaining below a threshold value.

19. An apparatus for selectively processing a digital image, said apparatus comprising:

a) an initial segmentation processor for performing an initial segmentation process on an image;

b) process performance data for a refined segmentation process wherein said performance data relates to image complexity;

c) an estimator for estimating a complexity measure for said image based on the results of said initial segmentation and said performance data; and

d) a refined segmentation processor for performing a refined segmentation process when said complexity measure meets a criterion; and

e) a segmentation-based enhancement processor for performing segmentation-based image enhancement based on said refined segmentation process when said complexity measure meets said criterion.

20. A method as described in claim 19 wherein said criterion comprises said complexity measure remaining below a threshold value.

21. A method for processing of a digital image, said method comprising:

a) performing an initial segmentation process on a non-object-based image data set;

b) estimating an initial complexity measure for said image data set based on the results of said initial segmentation process;

c) performing a refined segmentation process when said initial complexity measure meets an initial measure criterion;

d) estimating a refined complexity measure for said image data set based on the results of said refined segmentation process; and

e) performing an image enhancement process based on said refined segmentation process when said refined complexity measure meets a refined measure criterion.

22. A method as described in claim 21 further comprising performing a global image enhancement process when said refined complexity measure does not meet said refined measure criterion.

23. A method as described in claim 21 further comprising performing a global image enhancement process when said initial complexity measure does not meet said initial measure criterion.

24. A method as described in claim 21 further comprising performing a global image enhancement process when said refined complexity measure meets an alternative refined measure criterion.

25. A method as described in claim 21 further comprising performing a global image enhancement process when said initial complexity measure meets an alternative initial measure criterion.

26. A method for processing of a digital image, said method comprising:

a) performing an initial segmentation process on an image;

b) obtaining process performance data for a refined segmentation process wherein said performance data relates to image complexity;

c) estimating an initial complexity measure for said image based on the results of said initial segmentation process and said performance data;

d) performing a refined segmentation process when said initial complexity measure meets an initial measure criterion;

e) estimating a refined complexity measure for said image based on the results of said refined segmentation process; and

f) performing an image enhancement process based on said refined segmentation process when said refined complexity measure meets a refined measure criterion.

27. A method as described in claim 26 further comprising performing a global image enhancement process when said refined complexity measure does not meet said refined measure criterion.

28. A method as described in claim 26 further comprising performing a global image enhancement process when said initial complexity measure does not meet said initial measure criterion.

29. A method for processing of a digital image, said method comprising:

a) performing an initial segmentation process on a non-object-based image data set;

b) obtaining process performance data for a refined segmentation process wherein said performance data relates to image complexity;

c) estimating an initial complexity measure for said image data set based on the results of said initial segmentation process and said performance data;

d) performing a refined segmentation process when said initial complexity measure meets an initial measure criterion;

e) estimating a refined complexity measure for said image data set based on the results of said refined segmentation process; and

f) performing an image enhancement process based on said refined segmentation process when said refined complexity measure meets a refined measure criterion.

30. A method as described in claim 29 further comprising performing a global image enhancement process when said refined complexity measure does not meet said refined measure criterion.

31. A method as described in claim 29 further comprising performing a global image enhancement process when said initial complexity measure does not meet said initial measure criterion.