METHOD OF DETECTING REDEYE IN A DIGITAL IMAGE

• US 20080112599A1
• Filed: 11/07/2007
• Published: 05/15/2008
• Est. Priority Date: 11/10/2006
• Status: Active Grant

• Current Assignee: FotoNation Limited (Adeia Inc.)
• Original Assignee: FotoNation Vision Limited (Adeia Inc.)
• Inventors: Steinberg, Eran, Capata, Adrian, Nanu, Florin, Drimbarean, Alexandru, Corcoran, Peter
• Granted Patent: US 8,170,294 B2
• Time in Patent Office: Days
• US Class Current: 382/117
• CPC Class Codes: G06T 2207/30216 Redeye defect
• CPC Class Codes: G06T 5/005 Retouching; Inpainting; Scratch removal detecting, correction, reducing or removing defects, e.g. non-responsive pixels of solid state image sensors H04N25/68, scratch removal for cinematographic films scanned by electronic image sensor H04N5/253
• CPC Class Codes: G06T 5/77 Retouching; Inpainting; Scratch removal
• CPC Class Codes: G06T 7/90 Determination of colour characteristics
• CPC Class Codes: G06V 40/171 Local features and components; Facial parts eye characteristics G06V40/18; Occluding parts, e.g. glasses; Geometrical relationships
• 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 detecting redeye in a digital image comprises initially examining the image to detect redeyes, examining the image to detect face regions and, from the results of the preceding examinations, identifying those detected face regions each including only one detected redeye. Next, the identified face regions are examined, using less stringent search criteria than the initial examination, to detect additional redeyes in the face regions.

Claims

1. A method for detecting redeye in a digital image, comprising:

a) examining at least a portion of the image to detect redeye defects using first search criteria;

b) examining the image for detected face regions;

c) from the results of steps a) and b), identifying those detected face regions each having only one detected redeye defect; and

d) examining at least a portion of the face regions identified in c), using second search criteria to determine if a second redeye defect is present in one or more of said face regions.

2. A method as claimed in claim 1, further comprising:

e) combining the results of step a) and step d).

3. A method as claimed in claim 1, wherein said image comprises an image within a stream of images and wherein said face regions are detected in one or more images of said stream, said face regions being extrapolated from said one or more images to said image.

4. A method as claimed in claim 3 wherein the examining at least a portion of the image to detect redeye defects comprises examining said detected face regions for redeyes.

5. A method as claimed in claim 1 wherein said step of examining at least a portion of the image to detect redeye defects comprises examining substantially said complete image for redeye defects and wherein the examining at least a portion of the face regions comprises examining substantially said face region to determine if a second redeye defect is present.

6. The method claimed in claim 1, wherein c) further comprises identifying those detected face regions each having no detected redeye defects.

7. The method claimed in claim 1 wherein the second search criteria are less stringent than the first search criteria.

8. The method claimed in claim 1 wherein the first search criteria and the second search criteria each comprise a corresponding filter set.

9. The method claimed in claim 8 wherein the filter set corresponding to the second search criteria comprises less filters than the filter set corresponding to the first search criteria.

10. The method claimed in claim 7 wherein filter parameters of the filter set corresponding to the second search criteria are set broader than filter parameters of the filter set corresponding to the first search criteria.

11. The method claimed in claim 9 wherein the filter set corresponding to the second search criteria include filters arranged to detect non-redeye defects.

12. The method according to claim 11 further comprising:

detecting non-redeye defect pixels; and

correcting said non-redeye defect pixels by utilizing texture information derived from pixels of a detected redeye defect.

13. The method according to claim 12 wherein said texture information is derived from un-corrected pixels of said redeye defect.

14. The method according to claim 12 wherein said texture information is derived from corrected pixels of said redeye defect.

15. The method according to claim 12 wherein said texture information is derived from corrected pixels of a redeye defect located within the same face region as said second redeye defect.

16. The method according to claim 12, wherein said texture information comprises color information.

17. The method according to claim 11 wherein the identified portion of the face region includes a combined redeye and non-redeye defect, and wherein said texture information is derived from redeye pixels of said combined eye defect.

20. A computer program product comprising computer program code which when executed on a digital image acquisition apparatus is arranged to perform the method of claim 1 or claim 18.

21. A digital image processing device operable to perform the method of claim 1 or claim 18.

18. A method for detecting redeye in a digital image, comprising:

a) examining at least a portion of the image to detect redeye defects using a first search criteria;

b) examining the image for detected face regions;

c) from the results of a) and b), identifying each of those detected face regions having no detected redeye defects;

d) examining at least a portion of the face regions identified in c), using second search criteria to detect additional redeye defects in said face regions; and

e) combining the results from a) and d).

19. The method claimed in claim 18, wherein c) further comprises identifying each of those detected face regions having only one detected redeye defect.

22. A method for detecting pairs of redeye in a digital image, comprising:

a) examining at least a portion of the image to detect redeye defects using a first search criteria;

b) examining the image for detected face regions whereby said detected redeye defects are enclosed within the periphery of said detected face regions;

c) identifying said detected face regions each having only one detected redeye defect;

d) examining at least a portion of the face regions identified in c), using second search criteria, to detect a second redeye defect in one or more said face regions; and

e) combining the results from a) and d).

23. A method as recited in 22, further comprising:

determining the orientation of said detected face regions;

24. A method as recited in 23, wherein said second search criteria includes a search based on said determined orientation of said detected face regions.

25. A method as recited in 22 wherein said second search criteria includes anthropometric data relevant to said detected face regions and a determined location of a second eye based on a known location of said one detected redeye defect.