Method and apparatus for red-eye detection in an acquired digital image

• US 7,536,036 B2
• Filed: 10/28/2004
• Issued: 05/19/2009
• Est. Priority Date: 10/28/2004
• Status: Active Grant

• Current Assignee: FotoNation Limited (Adeia Inc.)
• Original Assignee: FotoNation Vision Limited (Adeia Inc.)
• Inventors: Biglol, Petronel, Steinberg, Eran, Prilutsky, Yury, Nanu, Florin, Corcoran, Peter
• Primary Examiner(s): Desire; Gregory M
• Application Number:
  US10/976,336
• Publication Number:
  US 20060093212A1
• Time in Patent Office: 1,664 Days
• Field of Search: 348/222.1, 348/239, 348/246, 348/333.01, 348/333.13, 348/372, 348/576, 358/3.26, 358/451, 358/518, 358/527, 382/115, 382/117, 382/162, 382/163, 382/164, 382/167, 382/168, 382/190, 382/246, 382/254, 382/274, 382/275, 382/165, 382/209, 382/260
• US Class Current: 382/117
• CPC Class Codes: G06T 2207/30216 Redeye defect
• CPC Class Codes: G06T 5/77 Retouching; Inpainting; Scratch removal
• CPC Class Codes: G06T 5/94 based on local image properties, e.g. for local contrast enhancement
• CPC Class Codes: G06T 7/90 Determination of colour characteristics
• CPC Class Codes: G06V 40/193 Preprocessing; Feature extraction
• CPC Class Codes: H04N 1/62 Retouching, i.e. modification of isolated colours only or in isolated picture areas only
• CPC Class Codes: H04N 1/624 Red-eye correction control of camera based on recognised facial parts H04N23/611

Abstract

A method for red-eye detection in an acquired digital image comprises acquiring a first image and analyzing the first acquired image to provide characteristics indicative of image quality. The process then determines if one or more corrective processes can be beneficially applied to the first acquired image according to the characteristics. Any such corrective processes are then applied to the first acquired image. Red-eye defects are then detected in a second acquired image using the corrected first acquired image. Defect detection can comprise applying one or more, or a chain of two or more, red-eye filters to the first acquired image. In this case, prior to the detecting step, it is determined if the red-eye filter or red eye filter chain can be adapted in accordance with the characteristics. The red-eye filter may be adapted accordingly.

Claims

1. A method for red-eye detection in an acquired digital image comprising:

a) acquiring a first image using a processor to perform the method of;

b) analyzing the first acquired image to provide one or more characteristics indicative of image quality;

c) determining if one or more corrective processes can be beneficially applied to said first acquired image according to said characteristics;

d) applying any such corrective processes to said first acquired image, including applying a chain of two or more red-eye filters to said first acquired image; and

e) detecting red-eye defects in a second acquired image using said corrected first acquired image.

2. A method according to claim 1 further comprising, prior to said detecting:

f) determining if said red-eye filter chain can be adapted in accordance with said plurality of characteristics; and

g) adapting said red-eye filter chain accordingly.

3. A method according to claim 2 in which said adapting comprises providing an altered set of parameters for one or more filters of said filter chain.

4. A method according to claim 2 in which said adapting comprises re-ordering a sequence in which said filters are applied to said first acquired image.

5. A method according to claim 2 in which said adapting comprises determining which filters are to be applied in said chain.

6. A method according to claim 1 in which more than one of said filters are applied simultaneously.

7. A method according to claim 1, further comprising correcting a third acquired image based on any defects detected in said detecting.

8. A method according to claim 7, wherein said first second and third images are the same image.

9. A method according to claim 7, wherein at least one of said first and second images is a sub-sampled copy of said third image.

10. A method as claimed in claim 7, wherein said applying a corrective process includes interacting with a user to determine the corrections to be made to said third image.

11. A method according to claim 1, wherein said first and second acquired images are the same image.

12. A method according to claim 1, wherein said first and second acquired images are different images.

13. A method according to claim 12, wherein said first acquired image is a sub-sampled copy of an acquired image.

14. A method according to claim 12, wherein said first acquired image is a sub-sampled copy of the second image.

15. A method according to claim 1, further comprising:

prior to analyzing the first acquired image, processing said second image based on the settings of a device used to acquire the second image.

16. A method according to claim 1, wherein said analyzing comprises determining one or more of:

a degree of blur;

a degree of dust contamination;

color balance;

white balance;

a gamma correction which might be applied;

texture characteristics;

noise characteristics; and

characteristics of regions;

in the first acquired image.

17. A method according to claim 1 wherein said corrective processes comprise one or more of:

(i) contrast normalization and image sharpening;

(ii) image color adjustment and tone scaling;

(iii) exposure adjustment and digital fill flash;

(iv) brightness adjustment with color space matching;

(v) image auto-gamma determination with image enhancement;

(v) image enhancement; and

(vi) face based image enhancement.

18. A method according to claim 1 wherein said red-eye filter chain comprises:

(i) a pixel locator and segmentor;

(ii) a shape analyzer; and

(iii) a falsing analyzer.

19. A method according to claim 18, wherein said pixel locator and segmentor includes a pixel transformer.

20. A method as claimed in claim 1, further comprising, responsive to determining to perform corrective processing, and further determining that such corrective processing is inapplicable to said first acquired image according to said characteristics, disabling said detecting of red-eye defects and providing an indication of such to a user.

21. A method according to claim 1, wherein the applying of a corrective process includes performing a color space transformation.

22. A method as claimed in claim 1 in which one or more of (a) to (e) are performed in a digital camera.

23. A method as claimed in claim 1, further comprising determining a sequence in which more than one corrective process can be beneficially applied to said first acquired image according to said characteristics.

24. A method as claimed in claim 1, wherein said one or more characteristics comprise a plurality of at least two characteristics.

25. A computer-readable storage medium containing a set of instructions which, when executed on a digital image processing device, program the device to performs the method of claim 1.

26. A digital image processing device including a processor configured to perform the method of claim 1.

27. A method for red-eye detection in an acquired digital image, comprising:

a) acquiring a first image, using a processor to perform the method of;

b) analyzing the first acquired image to provide one or more characteristics indicative of image quality;

c) determining if one or more red-eye filters can be adapted in accordance with said characteristics;

d) adapting said one or more red-eye filters as adapted red eye filters; and

e) applying said one or more adapted red eye filters to detect red-eye defects in a second acquired image.

28. A method according to claim 27, further comprising:

f) determining if one or more corrective processes can be beneficially applied to said first acquired image according to said characteristics; and

g) applying any such corrective processes to said first acquired image.

29. A method according to claim 28, wherein said analyzing comprises determining one or more of:

a degree of blur;

a degree of dust contamination;

color balance;

white balance;

a gamma correction which might be applied;

texture characteristics;

noise characteristics; and

characteristics of regions;

in the first acquired image.

30. A method according to claim 28, wherein said corrective processes comprise one or more of:

(i) contrast normalization and image sharpening;

(ii) image color adjustment and tone scaling;

(iii) exposure adjustment and digital fill flash;

(iv) brightness adjustment with color space matching;

(v) image auto-gamma determination with image enhancement;

(v) image enhancement; and

(vi) face based image enhancement.

31. A method as claimed in claim 28, comprising, responsive to determining to perform corrective processing, and further determining that such corrective processing is inapplicable to said first acquired image according to said characteristics, disabling said detecting of red-eye defects and providing an indication of such to a user.

32. A method according to claim 28, wherein the applying of a corrective process includes performing a color space transformation.

33. A method as claimed in claim 28 comprising determining a sequence in which more than one corrective process can be beneficially applied to said first acquired image according to said characteristics.

34. A method according to claim 27, wherein said one or more characteristics comprise a plurality of at least two characteristics.

35. A method according to claim 27, wherein said one or more red eye filters comprise a chain of two or more red eye filters.

36. A method according to claim 35, in which said adapting comprises providing an altered set of parameters for one or more filters of said filter chain.

37. A method according to claim 35 in which said adapting comprises re-ordering a sequence in which said filters are applied to said first acquired image.

38. A method according to claim 35, in which more than one of said filters are applied simultaneously.

39. A method according to claim 35, in which said adapting comprises determining which filters are to be applied in said chain.

40. A method according to claim 35, wherein said red-eye filter chain comprises:

(i) a pixel locator and segmentor;

(ii) a shape analyser; and

(iii) a falsing analyser.

41. A method according to claim 40 wherein said pixel locator and segmentor includes a pixel transformer.

42. A method according to claim 27, further comprising correcting a third acquired image based on any defects detected in said detecting.

43. A method according to claim 42, wherein said first second and third images are the same image.

44. A method according to claim 42, wherein at least one of said first and second images is a sub-sampled copy of said third image.

45. A method as claimed in claim 42, wherein said applying of a corrective process includes interacting with a user to determine the corrections to be made to said third image.

46. A method according to claim 27, wherein said first and second acquired images are the same image.

47. A method according to claim 27, wherein said first and second acquired images are different images.

48. A method according to claim 47, wherein said first acquired image is a sub-sampled copy of an acquired image.

49. A method according to claim 47, wherein said first acquired image is a sub-sampled copy of the second image.

50. A method according to claim 27, further comprising:

prior to analyzing the first acquired image, processing said second image based on the settings of a device used to acquire the second image.

51. A method as claimed in claim 27, in which one or more of (a) to (e) are performed in a digital camera.

52. A computer-readable storage medium containing a set of instructions which, when executed on a digital image processing device, program the device to perform the method of claim 27.

53. A digital image processing device including a processor configured to perform the method of claim 27.