Method for segmenting a digital image into a foreground region and a key color region

• US 6,011,595 A
• Filed: 09/19/1997
• Issued: 01/04/2000
• Est. Priority Date: 09/19/1997
• Status: Expired due to Term

• Current Assignee: Apple Inc.
• Original Assignee: Eastman Kodak Company
• Inventors: Spaulding, Kevin E., Couwenhoven, Douglas W., Henderson, Todd R.
• Primary Examiner(s): Faile, Andrew I.
• Assistant Examiner(s): LE, UYEN T
• Application Number:
  US08/934,373
• Time in Patent Office: 837 Days
• Field of Search: 348/590, 348/586, 348/587, 348/592, 348/593, 348/239, 348/591, 348/652, 348/653, 348/584, 348/585, 348/598, 348/599, 382/164, 382/165, 382/173, 358/538, 358/464
• US Class Current: 348/590
• CPC Class Codes: H04N 9/75 Chroma key

Abstract

A method for segmenting a digital image into a foreground region and a key color region including characterizing a distribution of key color values; using the distribution of key color values to produce a multi-dimensional look-up-table having three or more dimensions indexed by color values in a selected color space to define a key color volume; addressing the multi-dimensional look-up table with color values of pixels in the digital image to determine whether the color values of such pixels are within the key color volume; and labeling pixels in the digital image whose color values fall within the key color volume to be within the key color region and labeling all other pixels to be within the foreground region whereby the digital image is segmented into the foreground region and the key color region.

Claims

1. A method for segmenting a digital image into a foreground region and a key color region comprising the steps of:

a) characterizing a distribution of key color values;

b) using the distribution of key color values to produce a multi-dimensional look-up table having three or more dimensions indexed by color values in a selected color space to defme a key color volume;

c) addressing the multi-dimensional look-up table with color values of pixels in the digital image to determine whether the color values of such pixels are within the key color volume; and

d) labeling pixels in the digital image whose color values fall within the key color volume to be within the key color region and labeling all other pixels to be within the foreground region whereby the digital image is segmented into the foreground region and the key color region.

2. The method of claim 1 wherein the digital image is obtained from a digital camera.

3. The method of claim 1 wherein the digital image is obtained from a video camera.

4. The method of claim 1 wherein the digital image is obtained from a digitized photograph.

5. The method of claim 1 wherein step a) includes determining the distribution of color values in a digital calibration image that contains no foreground objects.

6. The method of claim 5 where the digital calibration image includes a shadowed key color region.

7. The method of claim 1 wherein step a) includes determining the distribution of key color values by analyzing a subset of the key color region in the digital image.

8. The method of claim 7 wherein the key color region is identified by a user.

9. The method of claim 1 wherein step a) includes having a user adjust a set of parameter values that characterize the distribution of key color values.

10. The method of claim 1 wherein the pixels in the digital image are labeled by an output value of the multi-dimensional look-up table.

11. The method of claim 1 wherein the selected color space comprises a luminance value and two chrominance values.

12. The method of claim 11 wherein the color space is CIELAB.

13. The method of claim 11 wherein the color space is YCrCb.

14. The method of claim 1 wherein the key color volume is defined by the envelope of the distribution of the key color values.

15. The method of claim 1 wherein the key color volume is defined by enlarging the envelope of the distribution of key color values.

16. The method of claim 15 wherein the envelope of the distribution of key color values is enlarged to include color values corresponding to shadowed key color regions.

17. The method of claim 1 wherein a range of hue angles is used to define the key color volume.

18. The method of claim 17 wherein the range of hue angles is a function of a luminance signal.

19. The method of claim 1 wherein a range of chroma values is used to define the key color volume.

20. The method of claim 19 wherein the range of chroma values is a function of a luminance value.

21. A method for segmenting a digital image into a foreground region and a key color region comprising the steps of:

a) characterizing a distribution of key color values;

b) using the distribution of key color values to define a key color volume in a selected color space where the key color volume is characterized by a range of hue angles and a range of chroma values wherein at least one of such ranges is a function of a luminance value; and

c) labeling pixels in the digital image whose color values fall within the key color volume to be within the key color region and labeling all other pixels to be within the foreground region whereby the digital image is segmented into the foreground region and the key color region.

22. The method of claim 21 wherein step a) includes determining the distribution of color values in a digital calibration image that contains no foreground objects.

23. The method of claim 22 where the digital calibration image includes a shadowed key color region.

24. The method of claim 21 wherein step a) includes determining the distribution of key color values by analyzing a key color region in the digital image.

25. The method of claim 24 wherein the key color region is identified by a user.

26. The method of claim 21 wherein step a) includes having a user adjust a set of parameters that characterize the key color volume.

27. The method of claim 21 wherein the selected color space comprises a luminance value and two chrominance values.

28. The method of claim 27 wherein the color space is CIELAB.

29. The method of claim 27 wherein the color space is YCrCb.

30. The method of claim 21 wherein the key color volume is defined by the envelope of the distribution of the key color values.

31. The method of claim 21 wherein the key color volume is defined by enlarging the envelope of the distribution of key color values.

32. The method of claim 31 wherein the envelope of the distribution of key color values is enlarged to include color values corresponding to shadowed key color regions.

33. A method for segmenting a first digital image into a foreground region and a key color region and composing the foreground region of the first digital image with a background digital image to produce a composite digital image comprising the steps of:

a) characterizing a distribution of key color values;

b) using the distribution of key color values to produce a multi-dimensional look-up-table having three or more dimensions indexed by color values in a selected color space to define a key color volume;

c) addressing the multi-dimensional look-up table with color values of pixels in the digital image to determine whether the color values of such pixels are within the key color volume;

d) labeling pixels in the digital image whose color values fall within the key color volume to be within the key color region and labeling all other pixels to be within the foreground region whereby the digital image is segmented into the foreground region and the key color region; and

e) combining the first digital image with a background digital image using the segmented regions to produce the composite digital image.

34. A method for segmenting a first digital image into a foreground region and a key color region and composing the foreground region of the first digital image with a background digital image to produce a composite digital image comprising the steps of:

a) characterizing a distribution of key color values;

b) using the distribution of key color values to define a key color volume in a selected color space where the key color volume is characterized by a range of hue angles and a range of chroma values wherein at least one of such ranges is a function of a luminance value;

c) labeling pixels in the digital image whose color values fall within the key color volume to be within the key color region and labeling all other pixels to be within the foreground region whereby the digital image is segmented into the foreground region and the key color region; and

d) combining the first digital image with a background digital image using the segmented regions to produce the composite digital image.

35. A computer program product having a computer program stored therein for segmenting a digital image into a foreground region and a key color region comprising the steps of:

a) characterizing a distribution of key color values;

b) using the distribution of key color values to produce a multi-dimensional look-up-table having three or more dimensions indexed by color values in a selected color space to define a key color volume;

c) addressing the multi-dimensional look-up table with color values of pixels in the digital image to determine whether the color values of such pixels are within the key color volume; and

d) labeling pixels in the digital image whose color values fall within the key color volume to be within the key color region and labeling all other pixels to be within the foreground region whereby the digital image is segmented into the foreground region and the key color region.

36. A computer program product having a computer program stored therein for segmenting a digital image into a foreground region and a key color region comprising the steps of:

a) characterizing a distribution of key color values;

b) using the distribution of key color values to define a key color volume in a selected color space where the key color volume is characterized by a range of hue angles and a range of chroma values wherein at least one of such ranges is a function of a luminance value; and

c) labeling pixels in the digital image whose color values fall within the key color volume to be within the key color region and labeling all other pixels to be within the foreground region whereby the digital image is segmented into the foreground region and the key color region.

37. A computer program product having a computer program stored therein for segmenting a digital image into a foreground region and a key color region comprising the steps of:

a) characterizing a distribution of key color values;

b) using the distribution of key color values to produce a multi-dimensional look-up-table having three or more dimensions indexed by color values in a selected color space to define a key color volume;

c) addressing the multi-dimensional look-up table with color values of pixels in the digital image to determine whether the color values of such pixels are within the key color volume;

d) labeling pixels in the digital image whose color values fall within the key color volume to be within the key color region and labeling all other pixels to be within the foreground region whereby the digital image is segmented into the foreground region and the key color region; and

e) combining the first digital image with a background digital image using the segmented regions to produce the composite digital image.

38. A computer program product having a computer program stored therein for segmenting a digital image into a foreground region and a key color region comprising the steps of:

a) characterizing a distribution of key color values;

b) using the distribution of key color values to define a key color volume in a selected color space where the key color volume is characterized by a range of hue angles and a range of chroma values wherein at least one of such ranges is a function of a luminance value;

c) labeling pixels in the digital image whose color values fall within the key color volume to be within the key color region and labeling all other pixels to be within the foreground region whereby the digital image is segmented into the foreground region and the key color region; and

d) combining the first digital image with a background digital image using the segmented regions to produce the composite digital image.