Computer program product for redeye detection

• US 6,252,976 B1
• Filed: 05/26/1998
• Issued: 06/26/2001
• Est. Priority Date: 08/29/1997
• Status: Expired due to Term

• Current Assignee: Monument Peak Ventures, LLC (Dominion Harbor Enterprises, LLC)
• Original Assignee: Eastman Kodak Company
• Inventors: Schildkraut, Jay S., Gray, Robert T.
• Primary Examiner(s): Bella, Matthew C.
• Assistant Examiner(s): CHAWAN, SHEELA C
• Application Number:
  US09/084,774
• Time in Patent Office: 1,127 Days
• Field of Search: 382/117, 382/164, 382/209, 382/199, 382/167, 382/118
• US Class Current: 382/117
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/30201 Face
• CPC Class Codes: G06T 2207/30216 Redeye defect
• CPC Class Codes: G06T 7/0002 Inspection of images, e.g. flaw detection
• CPC Class Codes: G06T 7/90 Determination of colour characteristics
• CPC Class Codes: G06V 40/193 Preprocessing; Feature extraction
• CPC Class Codes: H04N 1/624 Red-eye correction control of camera based on recognised facial parts H04N23/611

Abstract

A computer program product for detecting eye color defects of a subject in an image due to flash illumination comprises: a computer readable storage medium having a computer program stored thereon for performing the steps of detecting skin colored regions in a digital image; searching the skin colored regions for groups of pixels with color characteristic of redeye defect; and correcting color of the pixels based on a location of redeye defect found in step (b).

Claims

1. A method for detecting eye color defects of a subject in an image due to flash illumination, the method comprising the steps of:

(a) detecting a skin colored region in a digital image;

(b) detecting a human face by determining if the skin colored region has a predetermined characteristic of a human face;

(c) detecting a pair of candidate redeye defects originating from a pair of eyes in or adjacent to the detected human face based on comparing a distance between the candidate redeye defects and a dimension of the detected human face; and

(d) selecting the candidate redeye defects as actual redeye defects based on the results of step (c).

2. The method as in claim 1, wherein step (c) includes providing a width of the detected human face as the dimension of the detected human face.

3. A method for detecting eye color defects of a subject in an image due to flash illumination, the method comprising the steps of:

(a) detecting a pair of candidate redeye defects originating from a pair of eyes in the image; and

(b) evaluating the pair of candidate redeye defects based on a characteristic of each defect in the pair in relation to a distance between the redeye defects.

4. The method as in claim 3, wherein step (b) includes providing a size of the redeye defect as the characteristic.

5. The method as in claim 3, wherein step (b) includes providing a match to an eye template as the characteristic.

6. A method for detecting and correcting eye color defects of a subject in an image due to flash illumination, the method comprising the steps of:

(a) detecting a skin colored region in a digital image;

(b) detecting a human face by determining if the skin colored region has a predetermined characteristic of a human face;

(c) detecting a pair of candidate redeye defects originating from a pair of eyes in or adjacent to the detected human face based on comparing a distance between the candidate redeye defects and a dimension of the detected human face;

(d) selecting the candidate redeye defects as actual redeye defects based on the results of step (c); and

(e) correcting the redeye defects.

7. The method as in claim 6, wherein step (c) includes providing a width of the detected human face as the dimension of the detected human face.

8. A method for detecting and correcting eye color defects of a subject in an image due to flash illumination, the method comprising the steps of:

(a) detecting a pair of candidate redeye defects originating from a pair of eyes in the image;

(b) evaluating the pair of candidate redeye defects based on a characteristic of each defect in the pair in relation to a distance between the redeye defects; and

(c) correcting the pair of redeye defects.

9. The method as in claim 8, wherein step (b) includes providing a size of the redeye defect as the characteristic.

10. The method as in claim 8, wherein step (b) includes providing a match to an eye template as the characteristic.

11. A method for detecting eye color defects of a subject in an image due to flash illumination, the method comprising the steps of:

(a) detecting a skin colored region in a digital image;

(b) detecting a human face by determining if the skin colored region has a predetermined characteristic of a human face;

(c) detecting a candidate redeye pixel based at least in part on a position of the candidate redeye pixel in a non-skin colored hole in the detected human face, wherein said detecting a candidate redeye pixel includes growing a candidate redeye defect from the candidate redeye pixel and evaluating the candidate redeye defect based on at least one of the size, color and shape of the candidate redeye defect grown from the candidate redeye pixel; and

(d) selecting a candidate redeye defect based on the results of step (c).

12. The method as claimed in claim 11 wherein step (c) further comprises the step of assigning a score based on at least one of the size, color and shape of the candidate redeye defect grown from the candidate redeye pixel.

13. The method as claimed in claim 12 further comprising the steps of scoring a pair of candidate redeye defects originating from a pair of eyes in or adjacent to the human face based on comparing the consistency of at least one of their size, shape and color;

and selecting the candidate redeye defects as actual redeye defects based on the scoring results.