Apparatus and a method for automatically detecting and reducing red-eye in a digital image

• US 6,278,491 B1
• Filed: 01/29/1998
• Issued: 08/21/2001
• Est. Priority Date: 01/29/1998
• Status: Expired due to Term

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Company (HP Inc.)
• Inventors: Wang, John Y. A., Zhang, HongJiang
• Primary Examiner(s): HO, TUAN V
• Application Number:
  US09/015,407
• Time in Patent Office: 1,300 Days
• Field of Search: H04/N.5222, 348/207, 348/222, 348/239, 348/370, 348/576, 348/2, 382/115, 382/117, 382/162, 382/164, 382/167, 382/168, 382/169, 382/171, 382/172, 382/291, 382/201, 382/224, 382/118, 396/18
• US Class Current: 348/370
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/30216 Redeye defect
• CPC Class Codes: G06T 5/77 Retouching; Inpainting; Scratch removal
• CPC Class Codes: G06T 7/90 Determination of colour characteristics
• CPC Class Codes: G06V 40/193 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

An automatic red-eye detection and reduction system is described. The automatic red-eye detection and reduction system includes a red-eye detector that detects if an image contains a red pupil without user intervention. The red-eye detector detects location and size of the red pupil if the image is detected to contain the red pupil. The automatic red-eye detection and reduction system also includes a red-eye reduction system that is coupled to the red-eye detector to change each red color pixel within the red pupil into a predetermined color such that color of the red pupil can be detected and changed without user intervention. A method of automatically detecting and reducing red-eye effect in a digital image is also described.

Claims

1. An automatic red-eye detection and reduction system, comprising:

a red-eye detector that detects if an image contains a red pupil without user intervention, wherein the red-eye detector detects location and size of the red pupil if the image is detected to contain the red pupil, wherein the red-eye detector further comprises

a face detector that detects if the image contains a face, wherein the face detector determines the location, size, and orientation of the face if the image is detected to contain the face;

a red pupil detector coupled to the face detector to determine if the face contains the red pupil, wherein the red pupil detector determines the location and size of the red pupil within the face if the face is detected to contain the red pupil;

a red-eye reduction system coupled to the red-eye detector to change each red color pixel within the red pupil into a predetermined color such that color of the red pupil can be detected and changed without user intervention.

2. The system of claim 1, wherein the face detector is one of a neural network-based face detector and a principle component analysis-based face detector.

3. The system of claim 1, further comprising a rotating module that rotates the image for a predetermined rotation angle before the image is sent to the face detector, wherein the orientation of the face is determined by the rotation angle of the image if the image is detected to contain the face.

4. An automatic red-eye detection and reduction system, comprising:

a red-eye detector that detects if an image contains a red pupil without user intervention, wherein the red-eye detector detects location and size of the red pupil if the image is detected to contain the red pupil, wherein the red-eye detector further comprises

an eye detector that detects if the image contains an eye, wherein the eye detector determines the location, size, and orientation of the eye within the image if the image is detected to contain the eye;

a red pupil detector coupled to the eye detector to determine if the eye contains the red pupil, wherein the red pupil detector determines the location and size of the red pupil within the eye if the eye is detected to contain the red pupil;

a red-eye reduction system coupled to the red-eye detector to change each red color pixel within the red pupil into a predetermined color such that color of the red pupil can be detected and changed without user intervention.

5. The system of claim 4, wherein the eye detector is one of a neural network-based eye detector and a principle component analysis-based eye detector, wherein the red pupil detector determines if the eye contains the red pupil by locating a substantially circular region within the eye that contains red color pixels.

6. The system of claim 4, further comprising a rotating module that rotates the image for a predetermined rotation angle before the image is sent to the eye detector, wherein the orientation of the eye is determined by the rotation angle of the image if the image is detected to contain the eye.

7. An automatic red-eye detection and reduction system, comprising:

a red-eye detector that detects if an image contains a red pupil without user intervention, wherein the red-eye detector detects location and size of the red pupil if the image is detected to contain the red pupil, wherein the red-eye detector further comprises

a face detector that detects if the image contains a face, wherein the face detector determines the location, size, and orientation of the face if the image is detected to contain the face;

an eye detector coupled to the face detect or to detect if the face contains an eye, wherein the eye detector determines the location, size, and orientation of the eye within the face if the face is detected to contain the eye;

a red pupil detector coupled to the eye detector to determine if the eye contains the red pupil, wherein the red pupil detector determines the location and size of the red pupil within the eye if the eye is detected to contain the red pupil;

a red-eye reduction system coupled to the red-eye detector to change each red color pixel within the red pupil into a predetermined color such that color of the red pupil can be detected and changed without user intervention.

8. The system of claim 7, wherein the face detector is one of a neural network-based face detector and a principle component analysis-based face detector, wherein the eye detector is one of a neural network-based eye detector and a principle component analysis-based eye detector.

9. The system of claim 7, wherein the red pupil detector determines if the eye contains the red pupil by determining a substantially circular region within the eye that contains red color pixels.

10. The system of claim 7, further comprising a rotating module that rotates the image for a predetermined rotation angle before the image is sent to the face detector, wherein the orientation of the face is determined by the rotation angle of the image if the image is detected to contain the face.

11. A method for automatically detecting and reducing red-eye effect in a digital image, comprising:

(A) detecting if an image contains a red pupil without user intervention, wherein the location and size of the red pupil is also determined if the image is detected to contain the red pupil, wherein the step (A) further comprises the steps of

(I) detecting if the image contains a face and, if so, the location, size, and orientation of the face;

(II) detecting if the face contains the red pupil and, if so, the location and size of the red pupil within the eye;

(B) changing each red color pixel within the red pupil into a predetermined color such that color of the red pupil ca n be detected and changed without user intervention.

12. The method of claim 11, wherein the step (A) further comprises the step of rotating the image for a predetermined rotation angle prior to the step (I), wherein the orientation of the face is determined by the rotation angle.

13. The method of claim 11, wherein the step (II) determines if the face contains the red pupil by determining a substantially circular region within the face that contains red color pixels.

14. The method of claim 11, wherein the step (A) further comprises the steps of

(I) detecting if the image contains a face and, if so, the location, size, and orientation of the face;

(II) determining if the face contains a substantially open eye and, if so, the location of the eye within the face

(III) determining if the eye contains the red pupil and, if so, the location and size of the red pupil within the eye.

15. The method of claim 14, wherein the step of (III) determines if the eye contains the red pupil by determining a substantially circular region within the eye that contains red color pixels.

17. A method for automatically detecting and reducing red-eye effect in a digital image, comprising:

(A) detecting if an image contains a red pupil without user intervention, wherein the location and size of the red pupil is also determined if the image is detected to contain the red pupil, wherein the step (A) further comprises the steps of

(I) detecting if the image contains an eye and, if so, the location, size, and orientation of the eye within the image;

(II) determining if the eye contains the red pupil and, if so, the location, size, and orientation of the red pupil within the eye;

(B) changing each red color pixel within the red pupil into a predetermined color such that color of the red pupil can be detected and changed without user intervention.

16. The method of claim 19, wherein the step (A) further comprises the step of rotating the image for a predetermined rotation angle prior to the step (I), wherein the orientation of the face is determined by the rotation angle.

18. The method of claim 17, wherein the step (II) determines if the eye contains the red pupil by determining a substantially circular region within the eye that contains red color pixels.

19. The method of claim 17, wherein the step (A) further comprises the step of rotating the image for a predetermined rotation angle prior to the step (I), wherein the orientation of the eye is determined by the rotation angle.