Digital image processing system and method for emphasizing a main subject of an image
View Patent ↗A system and method for automatically altering an image having pixels is disclosed. An image is segmented into regions and various features of the regions are extracted and used to calculate a confidence factor, or belief value, that a particular region is a main subject of the image. The belief values are assembled into a belief map that is used to define how pixel values are to be altered. Various color and appearance values of the pixel are altered in a predetermined way to emphasize the main subject of the image. The emphasis is applied in a binary fashion or as in a gradient of values. Either the main subject can be emphasized, or the secondary or background subjects can be de-emphasized. The system can be implement on a variety of computer systems.
1. A method of modifying an image having pixels, comprising the steps of:
generating a main subject belief map containing values indicating the location of a plurality of regions in the image;
determining a threshold belief value discriminating among said regions, a main subject region and background region;
altering pixel values in said plurality of regions of said image, said altering following said generating, said altering being different in said main subject region and said background region.
2. The method of claim 1 wherein said generating and altering steps are without manual intervention.
3. The method of claim 1 wherein said generating of said map is based upon degree of centrality, borderness, and chrominance space distribution.
4. The method of claim 1 wherein said main subject includes skin tone pixels.
5. The method of claim 1 wherein said computing further comprises:
calculating a level of saliency of each of said regions; and
assigning a belief value to each of said pixels corresponding to the level of saliency of the respective said region.
6. The method of claim 5 wherein said calculating further comprises considering a plurality of semantic features including skin regions and a plurality of structural features including centrality.
7. The method of claim 6 wherein said main subject has a high degree of said centrality.
8. The method of claim 6 wherein said main subject includes skin tone pixels.
9. The method of claim 1 wherein said altering further comprises:
determining a characteristic of each pixel; and
altering said characteristics according to said associated belief values;
wherein said characteristics include one or more of: a saturation value, a luminance value, a hue value, a blur value.
10. The method of claim 1 wherein said altering further comprises:
determining the luminance values for the pixels of said background region; and
replacing the color values of the pixels of said background region with respective luminance values.
11. The method of claim 1 wherein said altering further comprises enhancing saturation of the pixels of said main subject region.
12. The method of claim 1 wherein said altering further comprises inverting the pixel hue values of the pixels that are a part of said main subject region.
13. The method of claim 1 wherein said altering further comprises inverting the pixel hue values of the pixels that are part of said background region.
14. A method of modifying an image having pixels, comprising the steps of:
automatically generating a main subject belief map containing values indicating the location of a plurality of regions in the image;
determining a main subject threshold belief value and a background threshold belief value, said threshold belief values discriminating a main subject region, a secondary subject region, and a background region; and
automatically altering pixel values in said plurality of regions of said image, said altering following said generating, said altering being different in each of said main subject, secondary subject, and background regions.
15. The method of claim 14 wherein said generating and altering steps are without manual intervention.
16. The method of claim 14 wherein said generating of said map is based upon degree of centrality, borderness, and chrominance space distribution.
17. The method of claim 14 wherein said main subject includes skin tone pixels.
18. The method of claim 14 wherein said computing further comprises:
calculating a level of saliency of each of said regions; and
assigning a belief value to each of said pixels corresponding to the level of saliency of the respective said region.
19. The method of claim 18 wherein said calculating further comprises considering a plurality of semantic features including skin regions and a plurality of structural features including centrality.
20. The method of claim 19 wherein said main subject has a high degree of said centrality.
21. The method of claim 14 wherein said main subject includes skin tone pixels.
22. The method of claim 14 wherein said altering further comprises:
determining a characteristic value for each pixel; and
altering said characteristic values according to said associated belief values;
wherein said characteristic values include one or more of: a luminance value, a hue value, a blur value.
23. The method of claim 14 wherein said altering further comprises:
determining the luminance values for the pixels of said background region; and
replacing the color values of the pixels of said background region with respective luminance values.
24. The method of claim 14 wherein said altering further comprises enhancing saturation of the pixels of said main subject region.
25. The method of claim 14 wherein said altering further comprises inverting the pixel hue values of the pixels that are a part of said main subject region.
26. The method of claim 14 wherein said altering further comprises inverting the pixel hue values of the pixels that are part of said background region.
27. A method of modifying an input image having pixels, comprising the steps of:
receiving a user selection of one of a plurality of different emphasis types to provide a selected emphasis;
automatically generating a main subject belief map containing values indicating the location of a plurality of regions in the input image;
determining a main subject threshold belief value and a background threshold belief value, said threshold belief values discriminating a main subject region, a secondary subject region, and a background region; and
automatically altering pixel values in said plurality of regions of said input image to provide an output image, said altering following said generating, said altering being different in each of said main subject, secondary subject, and background regions, said altering being in accordance with said selected emphasis.
28. The method of claim 27 further comprising delivering said output image to an output device.
29. The method of claim 28 wherein said output device is a printer or a display.