Color region based recognition of unidentified objects

• US 6,393,147 B2
• Filed: 04/13/1998
• Issued: 05/21/2002
• Est. Priority Date: 04/13/1998
• Status: Expired due to Fees

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Sampat, Ketan R., Danneels, Gunner D.
• Primary Examiner(s): Boudreau, Leo
• Assistant Examiner(s): SHERALI, ISHRAT I
• Application Number:
  US09/059,641
• Publication Number:
  US 20010014173A1
• Time in Patent Office: 1,499 Days
• Field of Search: 382/165, 382/162, 382/168, 382/103, 382/305, 382/219, 382/107, 382/153, 382/224, 707/1-10, 707/506, 707/528, 358/515, 345/150, 345/431, 345/835, 345/838, 345/968, 345/783, 345/589, 356/326, 356/406
• US Class Current: 382/165
• CPC Class Codes: G06V 10/56 relating to colour

Abstract

A machine implemented method is disclosed. The method includes characterizing an object by color regions, and then identifying the object in accordance with at least the color region based characterization of the object. In one embodiment, the method further includes generating output response, such as audio response, in accordance with the identification result.

Claims

1. A method comprising:

capturing an image of each of a plurality of known objects;

characterizing each image of the plurality of known objects by dividing each image into a plurality of color regions;

capturing an image of an unidentified object;

characterizing the image of the unidentified object by dividing the image of the unidentified object into a plurality of color regions; and

identifying the unidentified object at least in part by;

comparing the color regions of the image of the unidentified object with the color regions of one or more of the images of the plurality of known objects;

comparing the number of color regions of each color in the image of the unidentified object to the number of color regions of each color in the image of the first known object;

comparing the size of each color region of each color in the image of the unidentified object to the size of each color region of each color in the image of the first known object; and

comparing the relative orientation of each color region of each color in the image of the unidentified object to the relative orientation of each color region of each color in the image of the first known object.

2. The method of claim 1, wherein identifying the unidentified object further comprises rejecting the first known object as the identity of the unidentified object if the number of color regions, the size of color regions, or the relative orientation of color regions of the image of the unidentified object and the image of the first known object do not match within predetermined tolerances.

3. The method of claim 1, wherein characterizing the image of the unidentified object by dividing the image into a plurality of color regions comprises:

identifying the color of each of a plurality of pixels of the image; and

assigning each of the plurality of pixels of the image to one of the plurality of color regions of the image.

4. The method of claim 3, wherein characterizing the image of the unidentified object by dividing the image into a plurality of color regions further comprises:

selecting a first pixel of the plurality of pixels of the image and assigning the first pixel to a first color region;

establishing the color of the first pixel as the color of the first color region;

establishing the coordinates of the first pixel as the reference coordinates of the first color region;

selecting a second pixel of the plurality of pixels of the image, wherein the second pixel is adjacent to the first pixel;

if the second pixel is the same color as the first pixel, assigning the second pixel to the first color region; and

if the color of the second pixel is different than the color of the first pixel;

assigning the second pixel to a second color region,

establishing the color of the second pixel as the color of the second color region, and

establishing the coordinates of the second pixel as the reference coordinates of the second color region.

5. The method of claim 4, wherein characterizing the image of the unidentified object by dividing the image into a plurality of color regions further comprises examining each of the remaining pixels of the plurality of pixels of the image and assigning each remaining pixel either to a color region of an adjacent pixel or to a new color region.

6. The method of claim 1, further comprising generating digital image data for the unidentified object.

7. The method of claim 1, wherein the unidentified object comprises a three-dimensional object.

8. The method of claim 1, wherein capturing the image of the unidentified object comprises capturing the image of the unidentified object as a frame of video signals.

9. The method of claim 1, further comprising generating an output response in accordance with the result of identifying the unidentified object.

10. A method comprising:

capturing an image of each of a plurality of known objects;

characterizing each image of the plurality of known objects by dividing each image into a plurality of color regions;

capturing an image of an unidentified object;

characterizing the image of the unidentified object by dividing the image of the unidentified object into a plurality of color regions, comprising;

identifying the color of each of a plurality of pixels of the image,

assigning each of the plurality of pixels of the image to one of the plurality of color regions of the image,

selecting a first pixel of the plurality of pixels of the image and assigning the first pixel to a first color region,

establishing the color of the first pixel as the color of the first color region,

establishing the coordinates of the first pixel as the reference coordinates of the first color region,

selecting a second pixel of the plurality of pixels of the image, wherein the second pixel is adjacent to the first pixel,

if the second pixel is the same color as the first pixel, assigning the second pixel to the first color region, and

if the color of the second pixel is different than the color of the first pixel;

assigning the second pixel to a second color region,

establishing the color of the second pixel as the color of the second color region, and

establishing the coordinates of the second pixel as the reference coordinates of the second color region; and

identifying the unidentified object at least in part by comparing the color regions of the image of the unidentified object with the color regions of one or more of the images of the plurality of known objects.

11. The method of claim 10, wherein characterizing the image of the unidentified object by dividing the image into a plurality of color regions further comprises examining each of the remaining pixels of the plurality of pixels of the image and assigning each remaining pixel either to a color region of an adjacent pixel or to a new color region.

12. The method of claim 10, further comprising generating digital image data for the unidentified object.

13. The method of claim 10, wherein the unidentified object comprises a three-dimensional object.

14. The method of claim 10, wherein capturing the image of the unidentified object comprises capturing the image of the unidentified object as a frame of video signals.

15. The method of claim 10, further comprising generating an output response in accordance with the result of identifying the unidentified object.

16. An object recognition apparatus comprising:

an image capturing device to capture the image of an unidentified object;

a memory having stored therein data representing a plurality of images of known objects, wherein the data includes data regarding the characterization of each image of the plurality of known objects obtained by dividing the image into a plurality of color regions; and

a processor, wherein the operations of the processor include;

characterizing the image of the unidentified object by dividing the image of the unidentified object into a plurality of color regions, and

identifying the unidentified object at least in part by;

comparing the color regions of the image of the unidentified object with the color regions of one or more of the images of the plurality of known objects;

comparing the number of color regions of each color in the image of the unidentified object to the number of color regions of each color in the image of the first known object;

comparing the size of each color region of each color in the image of the unidentified object to the size of each color region of each color in the image of the first known object; and

comparing the relative orientation of each color region of each color in the image of the unidentified object to the relative orientation of each color region of each color in the image of the first known object.

17. The apparatus of claim 16, wherein the operation of the processor identifying the unidentified object further comprises rejecting the first known object as the identity of the unidentified object if the number of color regions, the size of color regions, or the relative orientation of color regions of the image of the unidentified object and the image of the first known object do not match within predetermined tolerances.

18. The apparatus of claim 16, wherein the operation of the processor characterizing the unidentified image by dividing the image into a plurality of color regions comprises:

identifying the color of each of a plurality of pixels of the image; and

assigning each of the plurality of pixels of the image to one of the plurality of color regions of the image.

19. The apparatus of claim 18, wherein the operation of the processor characterizing the unidentified image by dividing the image into a plurality of color regions further comprises:

selecting a first pixel of the plurality of pixels of the image and assigning the first pixel to a first color region;

establishing the color of the first pixel as the color of the first color region;

establishing the coordinates of the first pixel as the reference coordinates of the first color region;

selecting a second pixel of the plurality of pixels of the image, wherein the second pixel is adjacent to the first pixel;

if the second pixel is the same color as the first pixel, assigning the second pixel to the first color region; and

if the color of the second pixel is different than the color of the first pixel;

assigning the second pixel to a second color region,

establishing the color of the second pixel as the color of the second color region, and

establishing the coordinates of the second pixel as the reference coordinates of the second color region.

20. The apparatus of claim 19, wherein the operation of the processor characterizing the unidentified image by dividing the image into a plurality of color regions further comprises examining each of the remaining pixels of the plurality of pixels of the image and assigning each remaining pixel either to a color region of an adjacent pixel or to a new color region.

21. The apparatus of claim 16, wherein the operations of the processor further comprise generating digital image data of the unidentified object.

22. The apparatus of claim 16, wherein the unidentified object comprises a three-dimensional object.

23. The apparatus of claim 16, wherein the image capturing device captures the image of the unidentified object as a frame of video signals.

24. The apparatus of claim 16, wherein the operations of the processor further comprise generating an output response in accordance with the result of identifying the unidentified object.

25. An object recognition apparatus comprising:

an image capturing device to capture the image of an unidentified object;

a memory having stored therein data representing a plurality of images of known objects, wherein the data includes data regarding the characterization of each image of the plurality of known objects obtained by dividing the image into a plurality of color regions; and

a processor, wherein the operations of the processor include;

characterizing the image of the unidentified object by dividing the image of the unidentified object into a plurality of color regions, comprising;

identifying the color of each of a plurality of pixels of the image,

assigning each of the plurality of pixels of the image to one of the plurality of color regions of the image,

selecting a first pixel of the plurality of pixels of the image and assigning the first pixel to a first color region;

establishing the color of the first pixel as the color of the first color region;

establishing the coordinates of the first pixel as the reference coordinates of the first color region;

selecting a second pixel of the plurality of pixels of the image, wherein the second pixel is adjacent to the first pixel;

if the second pixel is the same color as the first pixel, assigning the second pixel to the first color region; and

if the color of the second pixel is different than the color of the first pixel;

assigning the second pixel to a second color region,

establishing the color of the second pixel as the color of the second color region, and

establishing the coordinates of the second pixel as the reference coordinates of the second color region; and

identifying the unidentified object at least in part by comparing the color regions of the image of the unidentified object with the color regions of one or more of the images of the plurality of known objects.

26. The apparatus of claim 25, wherein the operation of the processor characterizing the unidentified image by dividing the image into a plurality of color regions further comprises examining each of the remaining pixels of the plurality of pixels of the image and assigning each remaining pixel either to a color region of an adjacent pixel or to a new color region.

27. The apparatus of claim 25, wherein the operations of the processor further comprise generating digital image data of the unidentified object.

28. The apparatus of claim 25, wherein the unidentified object comprises a three-dimensional object.

29. The apparatus of claim 25, wherein the image capturing device captures the image of the unidentified object as a frame of video signals.

30. The apparatus of claim 25, wherein the operations of the processor further comprise generating an output response in accordance with the result of identifying the unidentified object.

31. A machine-readable medium having stored thereon data representing sequences of instructions that, when executed by a processor, cause the processor to perform operations comprising:

capturing an image of each of a plurality of known objects;

characterizing each image of the plurality of known objects by dividing each image into a plurality of color regions;

capturing an image of an unidentified object;

characterizing the image of the unidentified object by dividing the image of the unidentified object into a plurality of color regions; and

identifying the unidentified object at least in part by;

comparing the color regions of the image of the unidentified object with the color regions of one or more of the images of the plurality of known objects,

comparing the number of color regions of each color in the image of the unidentified object to the number of color regions of each color in the image of the first known object;

comparing the size of each color region of each color in the image of the unidentified object to the size of each color region of each color in the image of the first known object; and

comparing the relative orientation of each color region of each color in the image of the unidentified object to the relative orientation of each color region of each color in the image of the first known object.

32. The medium of claim 31, wherein identifying the unidentified object further comprises rejecting the first known object as the identity of the unidentified object if the number of color regions, the size of color regions, or the relative orientation of color regions of the image of the unidentified object and the image of the first known object do not match within predetermined tolerances.

33. The medium of claim 31, wherein characterizing the image of the unidentified object by dividing the image into a plurality of color regions comprises:

identifying the color of each of a plurality of pixels of the image; and

assigning each of the plurality of pixels of the image to one of the plurality of color regions of the image.

34. The medium of claim 33, wherein characterizing the image of the unidentified object by dividing the image into a plurality of color regions further comprises:

selecting a first pixel of the plurality of pixels of the image and assigning the first pixel to a first color region;

establishing the color of the first pixel as the color of the first color region;

establishing the coordinates of the first pixel as the reference coordinates of the first color region;

selecting a second pixel of the plurality of pixels of the image, wherein the second pixel is adjacent to the first pixel;

if the second pixel is the same color as the first pixel, assigning the second pixel to the first color region; and

if the color of the second pixel is different than the color of the first pixel;

assigning the second pixel to a second color region,

establishing the color of the second pixel as the color of the second color region, and

establishing the coordinates of the second pixel as the reference coordinates of the second color region.

35. The medium of claim 34, herein characterizing the image of the unidentified object by dividing the image into a plurality of color regions further comprises examining each of the remaining pixels of the plurality of pixels of the image and assigning each remaining pixel either to a color region of an adjacent pixel or to a new color region.

36. The medium of claim 31, wherein the data further comprises instructions causing the processor to perform operations comprising obtaining digital image data for the unidentified object.

37. The medium of claim 31, wherein the unidentified object comprises a three-dimensional object.

38. The medium of claim 31, wherein capturing the image of the unidentified object comprises capturing the image of the unidentified object as a frame of video signals.

39. The medium of claim 31, the data further comprises instructions causing the processor to perform operations comprising generating an output response in accordance with the result of identifying the unidentified object.

40. A machine-readable medium having stored thereon data representing sequences of instructions that, when executed by a processor, cause the processor to perform operations comprising:

capturing an image of each of a plurality of known objects;

characterizing each image of the plurality of known objects by dividing each image into a plurality of color regions;

capturing an image of an unidentified object;

characterizing the image of the unidentified object by dividing the image of the unidentified object into a plurality of color regions, comprising;

identifying the color of each of a plurality of pixels of the image,

assigning each of the plurality of pixels of the image to one of the plurality of color regions of the image,

selecting a first pixel of the plurality of pixels of the image and assigning the first pixel to a first color region;

establishing the color of the first pixel as the color of the first color region;

establishing the coordinates of the first pixel as the reference coordinates of the first color region;

selecting a second pixel of the plurality of pixels of the image, wherein the second pixel is adjacent to the first pixel;

if the second pixel is the same color as the first pixel, assigning the second pixel to the first color region; and

if the color of the second pixel is different than the color of the first pixel;

assigning the second pixel to a second color region,

establishing the color of the second pixel as the color of the second color region, and

establishing the coordinates of the second pixel as the reference coordinates of the second color region; and

identifying the unidentified object at least in part by comparing the color regions of the image of the unidentified object with the color regions of one or more of the images of the plurality of known objects.

41. The medium of claim 40, wherein characterizing the image of the unidentified object by dividing the image into a plurality of color regions further comprises examining each of the remaining pixels of the plurality of pixels of the image and assigning each remaining pixel either to a color region of an adjacent pixel or to a new color region.

42. The medium of claim 40, wherein the data further comprises instructions causing the processor to perform operations comprising obtaining digital image data for the unidentified object.

43. The medium of claim 40, wherein the unidentified object comprises a three-dimensional object.

44. The medium of claim 40, wherein capturing the image of the unidentified object comprises capturing the image of the unidentified object as a frame of video signals.

45. The medium of claim 40, the data further comprises instructions causing the processor to perform operations comprising generating an output response in accordance with the result of identifying the unidentified object.