COLOR INDEXING FOR IMAGES

• US 20080044081A1
• Filed: 10/06/2006
• Published: 02/21/2008
• Est. Priority Date: 08/21/2006
• Status: Abandoned Application

• Current Assignee: Adam Lieb
• Original Assignee: Adam Lieb
• Inventors: Lieb, Adam
• Application Number:
  US11/539,624
• Publication Number:
  US 20080044081A1
• Time in Patent Office: Days
• US Class Current: 382/162
• CPC Class Codes: G06F 16/5838 using colour
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/90 Determination of colour characteristics

Abstract

In one embodiment, a method is presented. The method includes receiving image data for an image. The method further includes inventorying color values of the image data. The method also includes assigning color values of the image data to a set of predetermined precise color values. The method includes storing a list of precise color values with data indicating a source of the image data.

Claims

1. A method, comprising:

receiving image data for an image;

calculating precise color values of the image data;

assigning precise color values of the image data to a set of predetermined discreet color values based on a mapping algorithm; and

storing a list of precise color values and discreet color values with data indicating a source of the image data.

2. The method of claim 1, further comprising:

organizing pixels of the image into groups based on color values of the pixels, with each color value of a pixel in a group visually matching color values of other pixels in the group.

3. The method of claim 2, further comprising:

filtering out fleshtone color values from pixel groups of the image.

4. The method of claim 2, further comprising:

filtering out background color values from pixel groups of the image.

5. The method of claim 2, further comprising:

filtering out greyscale color values from pixel groups of the image.

6. The method of claim 2, further comprising:

removing low-scoring color values from pixel groups of the image.

7. The method of claim 2, further comprising:

identifying a solid border of the image; and

ignoring pixels contained in the solid border.

8. The method of claim 2, further comprising:

merging multiple similar color groups into a single color group.

9. The method of claim 8, further comprising:

assigning a score to each color group.

10. The method of claim 9, wherein:

assigning a score to each color group depends on a position score derived from the average likelihood that all the pixels in the group are in the background of the image.

11. The method of claim 10, wherein:

determining the likelihood that a pixel is in the background of the image includes checking an absolute position of each pixel in the image and whether each pixel is contained in an automatically identified background area.

12. The method of claim 11, wherein:

the score of each color group depends on the number of pixels from the image contained within the group.

13. The method of claim 9, wherein:

the score of each color group depends on the average likelihood that the pixels in the group represent a flesh tone color.

14. The method of claim 9, wherein:

the score of each color group depends on the average likelihood that the pixels in the group represent a greyscale color.

15. The method of claim 9, wherein:

the score of each color group depends on the presence or absence of other flesh tone, background, or greyscale colors in the image.

16. The method of claim 2, further comprising:

converting color groups into a single precise scored color value, based on the color of all pixels contained in the group and the score of the group.

17. The method of claim 16, further comprising:

removing low scoring precise colors.

18. The method of claim 17, further comprising:

identifying identical or near-identical images from the image set by comparing each image'"'"'s set of precise colors to every other image'"'"'s set of precise colors; and

removing identical or near-identical images from the image set.

19. The method of claim 1, further comprising:

assigning all discreet color values to one color in a set of pre-determined commonly-used colors.

20. The method of claim 19, wherein:

The discreet color value assignments are chosen by an operator.

21. The method of claim 20, wherein:

The mapping from precise color to discreet color is performed by a machine.

22. The method of claim 21, wherein:

The mapping uses the assignments of discreet colors to commonly used colors.

23. The method of claim 21, wherein:

mapping includes comparing the absolute difference between the precise color and discreet color

24. The method of claim 21, wherein:

mapping includes comparing the shading difference between the precise color and discreet color.

25. The method of claim 21, wherein:

mapping includes comparing the relative ratios of red, green, and blue between the precise color and discreet color.

26. The method of claim 21, wherein:

mapping depends on a match threshold of each discreet color.

27. The method of claim 26, wherein:

the match threshold is determined by a type of the discreet color type.

28. The method of claim 26, wherein:

the match threshold is determined by an operator.

29. The method of claim 1, further comprising:

classifying color values from the list of precise color values as primary colors and secondary colors within the list of precise color values for the given image; and

applying these classifications to the mapped discreet colors.

30. The method of claim 1, further comprising:

aggregating the list of discreet color values into aggregated data of discreet color values across the set of images.

31. The method of claim 30, further comprising:

determining the frequency of occurrence of each discreet color value in the aggregated data of discreet color values for the entire set of images.

32. The method of claim 1, further comprising:

limiting the list of discreet color values to a predetermined maximum number of entries; and

removing low-scoring color values from the list of precise color values.

33. The method of claim 32, further comprising:

filtering out fleshtone color values from the list of precise color values;

filtering out background color values from the list of precise color values; and

filtering out greyscale color values from the list of precise color values.

34. The method of claim 1, further comprising:

aggregating the list of discreet color values into aggregated data of discreet color values for a set of other images; and

comparing the list of discreet color values to aggregated data of discreet color values for a set of other images.

35. The method of claim 34, further comprising:

determining the most frequently occurring discreet color values in the aggregated data of discreet color values for a set of other images.

36. The method of claim 35, further comprising:

determining the most frequently selected discreet color values in the aggregated data of discreet color values for a set of other images.

37. The method of claim 1, further comprising:

searching the list of discreet color values for a user-specified discreet color value; and

matching the user-specified discreet color value to a discreet color value of the list of discreet color values.

38. The method of claim 37, further comprising:

retrieving the image associated with the list of discreet color values; and

presenting the image associated with the list of discreet color values to a user.

39. A method, comprising:

receiving image data for an image;

inventorying color values of the image data;

assigning color values of the image data to a set of predetermined discreet color values based on distance of a color value from a discreet color value to provide a list of discreet color values;

limiting the list of discreet color values to a predetermined maximum number of entries;

removing low-scoring color values from the list of discreet color values;

filtering out fleshtone color values from the list of discreet color values;

filtering out background color values from the list of discreet color values;

filtering out greyscale color values from the list of discreet color values.

classifying color values from the list of discreet color values as primary colors and secondary colors within the list of discreet color values.

storing a list of discreet color values with data indicating a source of the image data;

aggregating the list of discreet color values into aggregated data of discreet color values for a set of other images; and

comparing the list of discreet color values to aggregated data of discreet color values for a set of other images.

40. A machine-readable medium embodying instructions, the instructions, when executed by a processor, causing the processor to perform a method, the method, comprising:

receiving image data for an image;

inventorying color values of the image data;

assigning color values of the image data to a set of predetermined discreet color values based on distance of a color value from a discreet color value to provide a list of discreet color values; and

storing the list of discreet color values with data indicating a source of the image data.

41. The machine-readable medium of claim 40, wherein the method further comprises:

limiting the list of discreet color values to a predetermined maximum number of entries; and

removing low-scoring color values from the list of discreet color values.

42. The machine-readable medium of claim 41, wherein the method further comprises:

filtering out fleshtone color values from the list of discreet color values;

filtering out background color values from the list of discreet color values; and

filtering out greyscale color values from the list of discreet color values.

43. The machine-readable medium of claim 42, wherein the method further comprises:

classifying color values from the list of discreet color values as primary colors and secondary colors within the list of discreet color values.

44. The machine-readable medium of claim 40, wherein the method further comprises:

aggregating the list of discreet color values into aggregated data of discreet color values for a set of other images; and

comparing the list of discreet color values to aggregated data of discreet color values for a set of other images.

45. The machine-readable medium of claim 44, wherein the method further comprises:

determining the most frequently occurring discreet color values in the aggregated data of discreet color values for a set of other images.

46. The machine-readable medium of claim 45, wherein the method further comprises:

determining the most frequently selected discreet color values in the aggregated data of discreet color values for a set of other images.

47. The machine-readable medium of claim 40, wherein the method further comprises:

searching the list of discreet color values for a user-specified discreet color value; and

matching the user-specified discreet color value to a discreet color value of the list of discreet color values.

48. The machine-readable medium of claim 47, wherein the method further comprises:

retrieving the image associated with the list of discreet color values; and

presenting the image associated with the list of discreet color values to a user.

49. The machine-readable medium of claim 40, wherein the method further comprises:

limiting the list of discreet color values to a predetermined maximum number of entries;

removing low-scoring color values from the list of discreet color values.

filtering out fleshtone color values from the list of discreet color values;

filtering out background color values from the list of discreet color values;

filtering out greyscale color values from the list of discreet color values;

classifying color values from the list of discreet color values as primary colors and secondary colors within the list of discreet color values;

aggregating the list of discreet color values into aggregated data of discreet color values for a set of other images; and

comparing the list of discreet color values to aggregated data of discreet color values for a set of other images.

50. A system, comprising:

a processor;

a memory coupled to the processor;

a user interface coupled to the processor;

a network interface coupled to the processor;

wherein the system is to;

receive image data for an image;

inventory color values of the image data;

assign color values of the image data to a set of predetermined discreet color values based on a mapping algorithm to provide a list of discreet color values; and

store the list of discreet color values with data indicating a source of the image data.

51. The system of claim 50, wherein the system is further to:

search the list of discreet color values for a user-specified discreet color value; and

match the user-specified discreet color value to a discreet color value of the list of discreet color values.

52. The system of claim 51, wherein the system is further to:

retrieve the image associated with the list of discreet color values; and

present the image associated with the list of discreet color values to a user.

53. The system of claim 50, wherein the system is further to:

limit the list of discreet color values to a predetermined maximum number of entries; and

remove low-scoring color values from the list of discreet color values.

54. The system of claim 53, wherein the system is further to:

filter out fleshtone color values from the list of discreet color values;

filter out background color values from the list of discreet color values; and

filter out greyscale color values from the list of discreet color values.

55. The system of claim 54, wherein the system is further to:

classify color values from the list of discreet color values as primary colors and secondary colors within the list of discreet color values.

56. The system of claim 50, wherein the system is further to:

aggregate the list of discreet color values into aggregated data of discreet color values for a set of other images; and

compare the list of discreet color values to aggregated data of discreet color values for a set of other images.

57. The system of claim 56, wherein the system is further to:

determine the most frequently occurring discreet color values in the aggregated data of discreet color values for a set of other images.

58. The system of claim 57, wherein the system is further to:

determine the most frequently selected discreet color values in the aggregated data of discreet color values for a set of other images.

59. The system of claim 50, further comprising:

means for interacting with a database.