Automated removal of red eye effect from a digital image

• US 6,009,209 A
• Filed: 06/27/1997
• Issued: 12/28/1999
• Est. Priority Date: 06/27/1997
• Status: Expired due to Term

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Bien, David L., Acker, Kristin, Lawton, Daryl T.
• Primary Examiner(s): Lee, Thomas D
• Assistant Examiner(s): BRINICH, STEPHEN M
• Application Number:
  US08/884,207
• Time in Patent Office: 914 Days
• Field of Search: 382/274-276, 382/162-167, 358/518-523
• US Class Current: 382/275
• 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 apparatus and method is disclosed for automatically editing a digital color image to remove discoloration of the image, known as a "red eye" effect, by parsing the discoloration into regions and re-coloring the area of the discoloration based on the attributes of the discoloration. The editing process according to the invention automatically creates a bitmap that is a correction image, which is composited with the source image or a copy of it and displayed as the source image with the red eye effect corrected.

Claims

1. In a computer system, a method for automatically editing a color image to remove discoloration of the image caused by a red eye artifact, the method comprising the steps of:

identifying attributes of the red eye artifact;

defining regions of the red eye artifact based on the identified attributes;

selecting a color for re-coloring each region of the red eye artifact based on predetermined criteria for each of the regions; and

re-coloring each of the regions with the associated selected colors.

2. The method of claim 1 wherein the regions include a highlight region within the red eye artifact in which the discoloration is the brightest.

3. The method of claim 2 wherein the color of the highlight region includes a linear gradient of brightness of the color.

4. The method of claim 3 wherein the linear gradient progresses radially from a location within the highlight region that corresponds to a location of a spot in the red eye artifact that is the brightest spot in the artifact.

5. The method of claim 2 wherein the color of a region surrounding the highlight region includes a non-linear gradient of brightness that progresses radially outwardly with respect to the highlight region.

6. The method of claim 3 wherein a slope of the linear gradient is a function of a distance between the brightest and darkest spots in the red eye artifact.

7. The method of claim 1 including categories into which one or more of the regions are classified, where at least one of the categories defines color attributes for re-coloring a region classified in that category.

8. The method of claim 7 wherein one or more of the regions is divided into upper and lower portions based on the attributes of the red eye artifact and separately re-colored based on the classification of the region.

9. The method of claim 1 wherein the attributes and regions of the red eye artifact are stored in the memory as a mask that is colored and composited with the image to provide an edited image that includes the re-coloring, which corrects the discoloration caused by the red eye artifact.

10. The method of claim 1 wherein the criteria include a linear gradient of coloring in one region and a non-linear gradient in another region.

11. The method of claim 1 wherein the attributes include a location of a brightest spot within the red eye artifact.

12. The method of claim 1 wherein the image comprises pixels and the attributes include a distance each pixel is from the brightest spot in the red eye artifact.

13. The method of claim 1 wherein the regions of the red eye artifact include a highlight region, a peripheral region and inside and outside border regions.

14. A computer system for correcting a discoloration in a digital image, the computer system comprising:

a source image with a red eye artifact;

a mask registered with the source image and dividing the red eye artifact into regions;

a tool for coloring the mask, where the color for each region is at least in part selected based on attributes of the particular region of the mask in which the color is applied; and

, a composite image combining the source image and the colored mask, which visually appears as the source image with the red eye artifact replaced with a natural-looking coloring in an area of the pupil of the eye.

15. The computer system of claim 14 wherein the regions include a highlight region within the red eye artifact in which the discoloration is the brightest.

16. The computer system of claim 15 wherein the color of the highlight region includes a linear gradient of brightness of the color.

17. The computer system of claim 16 wherein the linear gradient progresses radially from a location within the highlight region that corresponds to a location of a spot in the red eye artifact that is the brightest spot in the artifact.

18. The computer system of claim 17 wherein the color of a region surrounding the highlight region includes a non-linear gradient of brightness that progresses radially outwardly with respect to the highlight region.

19. The computer system of claim 16 wherein a slope of the linear gradient is a function of a distance between the brightest and darkest spots in the red eye artifact.

20. The computer system of claim 14 including categories into which one or more of the regions are classified, where at least one of the categories defines color attributes for re-coloring a region classified in that category.

21. The computer system of claim 14 including attributes of the red eye artifact that are identified and stored in a memory as criteria for defining the regions.

22. In a computer for receiving and displaying a color image having a red eye artifact, an article of manufacture comprising a medium that is readable by the computer and that carries instructions for the computer to perform a process comprising the steps of:

parsing the red eye artifact into regions;

selecting a color for re-coloring the image in each of the regions of the red eye artifact based on attributes of the region; and

re-coloring each of the regions with the associated selected color.

23. The article of manufacture of claim 22 wherein the regions include a highlight region within which is the brightest coloring of the red eye artifact.

24. The article of manufacture of claim 22 including the step of identifying attributes of an area of the image that includes the red eye artifact;

and, defining the regions of the red eye artifact based on the identified attributes and storing the regions in the memory.

25. The article of manufacture of claim 22 wherein one of the attributes includes a location of the brightest spot in the red eye artifact.

26. A computer system for correcting a discoloration in a digital image, the computer system comprising:

a source image with a red eye artifact, a mask registered with the source image and dividing the red eye artifact into regions, wherein the regions include a highlight region within the red eye artifact in which the discoloration is the brightest, and wherein the color of the highlight region includes a linear gradient of brightness of the color;

a tool for coloring the mask, where the color is at least in part determined by the region of the mask in which the color is applied; and

, a composite image combining the source image and the colored mask, which visually appears as the source image with the red eye artifact replaced with a natural-looking coloring in an area of the pupil of the eye.