System and method for color characterization with applications in color measurement and color matching

• US 6,757,428 B1
• Filed: 08/17/1999
• Issued: 06/29/2004
• Est. Priority Date: 08/17/1999
• Status: Expired due to Term

• Current Assignee: National Instruments Corporation (Emerson Electric Company)
• Original Assignee: National Instruments Corporation (Emerson Electric Company)
• Inventors: Lin, Siming, Nair, Dinesh
• Primary Examiner(s): Do, Anh Hong
• Application Number:
  US09/375,453
• Time in Patent Office: 1,778 Days
• Field of Search: 382/165, 382/162, 382/164, 382/168, 382/167, 382/118, 382/107, 382/209, 382/219, 348/30, 348/187, 209/580, 209/587, 356/240.1
• US Class Current: 382/165
• CPC Class Codes: G06F 18/256 of results relating to different input data, e.g. multimodal recognition
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/30141 Printed circuit board [PCB]
• CPC Class Codes: G06T 7/001 using an image reference approach
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
• CPC Class Codes: G06V 10/56 relating to colour

Abstract

A color characterization method operates to analyze each respective pixel of at least a subset of the pixels of an image object. The image is obtained in HSI format, or alternatively converted from another format to HSI. For each respective pixel, the method determines a color category or bin for the respective pixel based on values of the respective pixel. The color category is one a plurality of possible color categories or bins in the HSI color space. As the pixels are analyzed and assigned to color categories, the method stores information in tho computer regarding the number or percentage or pixels in each of the color categories. A color matching method uses the color characterization method. The color matching method determines similarity of colors between a template image object and a region of interest (ROI). The color matching method first performs the above color characterization technique on each of a template image object and a ROI (region of interest) and then generates match information based on the color information of the template image object and the color information of the ROI, wherein this match information indicates the similarity of colors between the template image object and the ROI.

Claims

1. A computer-implemented method for performing color matching between a template image object and a region of interest (ROI), wherein each of the template image object and the ROI comprises a respective plurality of pixels, the method comprising:

for each respective pixel of at least a subset of the pixels of each of the template image object and the ROI, determining a color category for the respective pixel based on hue, saturation and intensity values of the respective pixel, wherein said determining determines a color category from a plurality of possible color categories;

storing information regarding the number of pixels in each of said plurality of possible color categories of each of each of the template image object and the ROI, wherein said information characterizes colors in the image object; and

generating match information based on the color information of the template image object and the color information of the ROI, wherein said match information indicates the similarity of colors between the template image object and the ROI.

2. The method of claim 1, wherein the method operates for all of the plurality of pixels in one or more of the template image object and the ROI.

3. The method of claim 1, wherein said plurality of possible color categories comprise color categories of a color space.

4. The method of claim 1, wherein the number of said color categories is dependent on a desired complexity of the color characterization.

5. The method of claim 1, further comprising:

receiving user input indicating a desired complexity of color characterization;

wherein the number of said color categories is dependent on the desired complexity of color characterization.

6. The method of claim 1, wherein said determining includes:

determining if the respective pixel can be categorized as either black or white based on one or more of saturation and intensity values of the respective pixel;

assigning the respective pixel to a black or white category if the respective pixel can be categorized as black or white, respectively;

determining a color category for the respective pixel based on hue and saturation values of the respective pixel if the respective pixel can not be categorized as either black or white.

7. The method of claim 1, further comprising;

performing a smoothing operation on said information after said storing.

8. The method of claim 7, wherein said smoothing operation comprises:

for each respective color category of at least a subset of said plurality of possible color categories, distributing a portion of the number of pixels in the respective color category to one or more neighboring color categories.

9. The method of claim 1, wherein said information includes percentage information of the percentage of pixels in each of said plurality of possible color categories, wherein said percentage information characterizes colors in one or more of the template image object and the ROI.

10. The method of claim 9, further comprising;

performing a smoothing operation on said percentage information after said storing;

wherein said smoothing operation comprises;

for each respective color category of at least a subset of said plurality of possible color categories, distributing a portion of the percentage of the number of pixels in the respective color category to one or more neighboring categories.

11. The method of claim 1, wherein the ROI comprises a portion of an image.

12. The method of claim 1, further comprising:

receiving one or more of the template image object and the ROI;

storing one or more of the template image object and the ROI in a memory of a computer;

wherein said receiving and said storing are performed prior to said determining a color category and said storing information.

13. The method of claim 1, further comprising:

converting one or more of the template image object and the ROI to hue, saturation, and intensity (HSI) format prior to said determining a color category and said storing information.

14. A computer-implemented method for performing color matching between a template image object and a region of interest (ROI), wherein each of the template image object and the ROI comprises a respective plurality of pixels, the method comprising:

analyzing each of at least a subset of the pixels, wherein, for each respective pixel, said analyzing comprises;

analyzing two or more of the hue, saturation, and intensity values of the respective pixel;

determining a color category for the respective pixel based on the two or more of the hue, saturation, and intensity values, wherein said determining determines a color category from a plurality of possible color categories;

storing information in the computer regarding the number of pixels in each of said plurality of possible color categories, wherein said information characterizes colors in each of the template image object and the ROI; and

generating match information based on the information of the template image object and the information of the ROI, wherein said match information indicates the similarity of colors between the template image object and the ROI.

15. A computer-implemented method for performing color matching between a template image object and a region of interest (ROI), wherein each of the template image object and the ROI comprises a respective plurality of pixels, wherein the method operates for at least a subset of the pixels, wherein, for each respective pixel of the at least a subset of the plurality of pixels, the method comprises:

analyzing two or more of the hue, saturation, and intensity values of the respective pixel;

determining a color category for the respective pixel based on the two or more of the hue, saturation, and intensity values, wherein said determining determines a color category from a plurality of possible color categories, thereby generating color information of the template image object and the ROI;

wherein the number of pixels in each of said color categories operates to characterize colors in each of the template image object and the ROI; and

generating match information based on the color information of the template image object and the color information of the ROI, wherein said match information indicates the similarity of colors between the template image object and the ROI.

16. The method of claim 15, further comprising:

storing information in the computer regarding the number of pixels in each of said plurality of possible color categories, wherein said information characterizes colors in one or more of the template image object and the ROI.

17. A computer-implemented method for performing color matching between a template image object and a region of interest (ROI), wherein each of the template image object and the ROI comprises a respective plurality of pixels, the method comprising:

for each of the template image object and the ROI,

for each respective pixel of at least a subset of the pixels, determining a color category for the respective pixel based on values of the respective pixel, wherein said determining determines a color category from a plurality of possible color categories, wherein said plurality of possible color categories comprise color categories of a color space;

storing information in the computer regarding the number of pixels in each of said plurality of possible color categories, wherein said information characterizes colors; and

generating match information based on the information of the template image object and the information of the ROI, wherein said match information indicates the similarity of colors between the template image object and the ROI.

18. The method of claim 17, wherein the method operates for all of the plurality of pixels in one or more of the template image object and the ROI.

19. The method of claim 17, wherein the number of said color categories is dependent on a desired complexity of the color characterization.

20. The method of claim 17, further comprising;

performing a smoothing operation on said information after said storing.

21. The method of claim 20, wherein said smoothing operation comprises:

for each respective color category of at least a subset of said plurality of possible color categories, distributing a portion of the number of pixels in the respective color category to one or more neighboring color categories.

22. The method of claim 17, wherein said storing information includes calculating percentage information of the number of pixels in each of said plurality of possible color categories, wherein said percentage information characterizes colors in one or more of the template image object and the ROI.

23. A memory medium comprising program instructions for determining similarity of colors between a template image object and a region of interest (ROI), wherein each of the template image object and the ROI comprises a respective plurality of pixels, wherein the program instructions are executable to implement:

for each of the template image object and the ROI,

for each respective pixel of at least a subset of the pixels, determining a color category for the respective pixel based on hue and saturation values of the respective pixel, wherein said determining determines a color category from a plurality of possible color categories;

storing information in the computer regarding the number of pixels in each of said plurality of possible color categories, wherein said information characterizes colors in; and

generating match information based on the color information of the template image object and the color information of the ROI, wherein said match information indicates the similarity of colors between the template image object and the ROI.

24. The memory medium of claim 23, wherein said plurality of possible color categories comprise color categories of a color space.

25. The memory medium of claim 23, wherein the number of said color categories is dependent on a desired complexity of the color characterization.

26. The memory medium of claim 23, further comprising;

performing a smoothing operation on said information after said storing.

27. A system for determining similarity of colors between a template image object and a region of interest (ROI), wherein each of the template image object and the ROI comprises a respective plurality of pixels, the system comprising:

a processor;

a memory medium coupled to the processor which stores each of the template image object and the ROI, wherein the memory medium also stores a color characterization program;

wherein the processor is operable to execute the color characterization program to;

for each of the template image object and the ROI,

for each respective pixel of at least a subset of the pixels, determine a color category for the respective pixel based on hue and saturation values of the respective pixel, wherein said determining determines a color category from a plurality of possible color categories;

store information in the memory medium regarding the number of pixels in each of said plurality of possible color categories, wherein said information characterizes colors in each of the template image object and the ROI; and

generating match information based on the color information of the template image object and the color information of the ROI, wherein said match information indicates the similarity of colors between the template image object and the ROI.

28. The system of claim 27, wherein said plurality of possible color categories comprise color categories of a color space.

29. The system of claim 27, wherein the number of said color categories is dependent on a desired complexity of the color characterization.

30. A method for determining similarity of colors between a template image object and a region of interest (ROI), comprising:

determining color information in the template image object, wherein the template image object comprises a plurality of template image object pixels, wherein said color information comprises information regarding a number of template image object pixels in each of a plurality of color categories;

determining color information in the ROI, wherein the ROI comprises a plurality of ROI pixels, wherein said color information comprises information regarding a number of ROI pixels in each of the plurality of color categories;

generating match information based on the color information of the template image object and the color information of the ROI, wherein said match information indicates the similarity of colors between the template image object and the ROI.

31. The method of claim 30, wherein said plurality of color categories comprise color categories of a color space.

32. The method of claim 30, wherein said plurality of color categories comprise color categories of the HSI color space.

33. The method of claim 30, wherein the number of said color categories is dependent on a desired complexity of a color characterization of the template image object and the ROI.

34. The method of claim 30, wherein said generating match information includes comparing the color information of the template image object and the color information of the ROI.

35. The method of claim 34, wherein said comparing includes, for each respective color category, comparing the number of template image object pixels in the respective color category with the number of ROI pixels in the respective color category.

36. The method of claim 34, wherein the color information of the template image object comprises information regarding a percentage of template image object pixels in each of the plurality of color categories;

wherein the color information of the ROI comprises information regarding a percentage of ROI pixels in each of the plurality of color categories;

wherein said comparing includes, for each respective color category, comparing the percentage of template image object pixels in the respective color category with the percentage of ROI pixels in the respective color category.

37. The method of claim 30, wherein the color information of the template image object comprises information regarding a percentage of template image object pixels in each of the plurality of color categories;

wherein the color information of the ROI comprises information regarding a percentage of ROI pixels in each of the plurality of color categories;

wherein said generating match information includes, for each respective color category, subtracting the percentage of template image object pixels in the respective color category from the percentage of ROI pixels in the respective color category.

38. The method of claim 30, wherein, for each of the template image object and the ROI, said determining color information comprises:

for each respective pixel of at least a subset of the pixels, determining a color category for the respective pixel based on hue and saturation values of the respective pixel, wherein said determining determines a color category from the plurality of color categories;

storing information in the computer regarding the number of pixels in each of said plurality of color categories.

39. The method of claim 38, further comprising;

performing a smoothing operation on said information after said storing.

40. The method of claim 39, wherein said smoothing operation comprises:

for each respective color category of at least a subset of said plurality of possible color categories, distributing a portion of the number of pixels in the respective color category to one or more neighboring color categories.

41. A memory medium comprising program instructions for determining similarity of colors between a template image object and a region of interest (ROI), wherein the program instructions are executable to implement:

determining color information in the template image object, wherein the template image object comprises a plurality of template image object pixels, wherein said color information comprises information regarding a number of template image object pixels in each of a plurality of color categories;

determining color information in the ROI, wherein the ROI comprises a plurality of ROI pixels, wherein said color information comprises information regarding a number of ROI pixels in each of the plurality of color categories;

generating match information based on the color information of the template image object and the color information of the ROI, wherein said match information indicates the similarity of colors between the template image object and the ROI.

42. A memory medium comprising program instructions for determining similarity of colors between a template image object and a region of interest (ROI),

wherein the memory medium stores color information corresponding to the template image object, wherein said color information comprises information regarding a number of template image object pixels in each of a plurality of color categories;

wherein the program instructions are executable to implement;

determining color information in the ROI, wherein the ROI comprises a plurality of ROI pixels, wherein said color information comprises information regarding a number of ROI pixels in each of the plurality of color categories;

generating match information based on the color information of the template image object and the color information of the ROI, wherein said match information indicates the similarity of colors between the template image object and the ROI.

43. A system for determining similarity of colors between a template image object and a region of interest (ROI), the system comprising:

a processor;

a memory medium coupled to the processor which stores color information corresponding to the template image object, wherein said color information comprises information regarding a number of template image object pixels in each of a plurality of color categories;

wherein the memory medium also stores a color matching program;

wherein the processor is operable to execute the color matching program to;

determine color information in the ROI, wherein the ROI comprises a plurality of ROI pixels, wherein said color information comprises information regarding a number of ROI pixels in each of the plurality of color categories;

generate match information based on the color information of the template image object and the color information of the ROI, wherein said match information indicates the similarity of colors between the template image object and the ROI.

44. A method for producing a more robust color characterization of an image object by sharing color information between neighboring color categories, comprising:

receiving said color information for an image object;

selecting a color category from the said received color information for the image object;

determining color categories to receive compensation from said selected color category;

determining a percentage of said selected color category to be distributed to said determined number of said neighboring color categories;

distributing said percentages to said number of neighboring color categories.

45. The method of claim 44, wherein the number of neighboring color categories to receive compensation is selected by the user.

46. The method of claim 44, wherein the number of neighboring color categories to receive compensation is automatically calculated according to the color information of the image object.