Representing an image with weighted joint histogram

• US 6,671,402 B1
• Filed: 12/28/2000
• Issued: 12/30/2003
• Est. Priority Date: 12/15/2000
• Status: Expired due to Term

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: America Online Inc. (Warner Bros. Discovery, Inc.)
• Inventors: Pass, Gregory S., Wood, Frank
• Primary Examiner(s): Boudreau, Leo
• Assistant Examiner(s): SHERALI, ISHRAT I
• Application Number:
  US09/749,626
• Time in Patent Office: 1,097 Days
• Field of Search: 382/305, 382/168, 382/170, 382/162, 382/165, 382/209, 382/218, 382/219, 382/220, 382/108, 382/224, 358/403, 358/518-522, 358/515-516, 348/403.1, 707/1-6, 707/7, 707/104.1, 345/700, 345/788, 345/419, 345/835, 345/584, 345/773, 345/8, 345/589, 345/597
• US Class Current: 382/170
• CPC Class Codes: G06F 16/5838 using colour
• CPC Class Codes: G06V 10/507 Summing image-intensity values; Histogram projection analysis

Abstract

Representing an image includes extracting several types of information about an image that are then used to compute a joint histogram. A weighting factor may be applied to at least one of the types of information extracted about the image such that the joint histogram represents a weighted joint histogram. Representing an image may further include calculating a posterized joint histogram. The posterized joint histogram may be calculated after the weighting factor has been applied.

Claims

1. A method for representing an image, the method comprising:

extracting more than one image characteristic from an image;

computing a joint histogram based on the more than one image characteristic extracted about the image;

determining a weighting factor based on at least one of a position of a pixel within the image, color, brightness, edge density, texturedness, gradient magnitude, centeredness, and rank; and

applying the weighting factor to at least one of the image characteristics extracted about the image by modifying a pixel count related to the image characteristic such that the joint histogram represents a weighted joint histogram.

2. The method as in claim 1 further comprising calculating a posterized joint histogram.

3. The method as in claim 1 further comprising calculating a posterized joint histogram after applying the weighting factor.

4. The method as in claim 1 wherein extracting the more than one image characteristic includes extracting information reflecting several features of the image.

5. The method as in claim 4 wherein extracting the more than one image characteristic includes extracting information regarding more than one of color, edge density, texturedness, gradient magnitude, and rank as related to pixels in the image.

6. The method as in claim 4 wherein:

computing the joint histogram includes computing a first histogram vector based on the information reflecting several features of the image, and

the first histogram vector includes one or more vector elements, each representing information for a different combination of features extracted about the image.

7. The method as in claim 6 further comprising calculating a posterized joint histogram that includes:

identifying multiple subsets within at least one of the vector elements included in the first histogram vector; and

creating a second histogram vector that includes a vector element for each of the identified subsets that are included in the first histogram vector, with data within the vector elements of the second histogram representing the image features extracted.

8. The method as in claim 1 wherein determining the weighting factor includes:

extracting one or more additional types of information about the image that differ from the image characteristics used to compute the joint histogram; and

determining whether to modify a vector element value in the joint histogram based on the additional information that was extracted about the image, with the joint histogram including one or more vector elements.

9. The method as in claim 8 wherein extracting one or more additional types of information includes extracting a centeredness feature about each pixel in the image, and determining whether to modify includes determining whether to modify the vector element values in the joint histogram based on the centeredness feature extracted about each pixel in the image.

10. The method as in claim 1 wherein applying the weighting factor includes applying the weighting factor as the joint histogram is being computed.

11. The method as in claim 1 wherein determining the weighting factor includes determining the weighting factor as the joint histogram is being computed.

12. The method as in claim 1 wherein determining the weighting factor includes determining the weighting factor based on a centeredness image characteristic extracted about the image.

13. The method as in claim 1 wherein the weighting factor is a non-zero weighting factor.

14. The method as in claim 1 wherein determining the weighting factor includes determining the weighting factor based on at least one image characteristic that differs from the image characteristics used to compute the joint histogram.

15. A computer program product comprising a computer readable medium on which a computer program is stored for representing an image, said program comprising:

an extracting code segment that causes the computer to extract more than one image characteristic from an image;

a computing code segment that causes the computer to compute a joint histogram based on the more than one image characteristic extracted about the image;

a determining code segment that causes the computer to determine a weighting factor based on at least one of a position of a pixel within the image, color, brightness, edge density, texturedness, gradient magnitude, centeredness, and rank; and

an applying code segment that causes the computer to apply the weighting factor to at least one of the image characteristics extracted about the image by modifying a pixel count related to the image characteristic such that the joint histogram represents a weighted joint histogram.

16. The computer program of claim 15 further comprising a calculating code segment that causes the computer to calculate a posterized joint histogram.

17. The computer program of claim 15 further comprising a calculating code segment that causes the computer to calculate a posterized joint histogram after applying the weighting factor.

18. The computer program of claim 15 wherein the extracting code segment includes a code segment that causes the computer to extract information reflecting more than one feature of the image.

19. The computer program of claim 18 wherein:

the computing code segment includes a code segment that causes the computer to compute a first histogram vector based on the information reflecting several features of the image, and

the first histogram vector includes one or more vector elements, each representing information for a different combination of features extracted about the image.

20. The computer program of claim 19 further comprising a calculating code segment that causes the computer to calculate a posterized joint histogram and that includes:

an identifying code segment that causes the computer to identify multiple subsets within at least one of the vector elements included in the first histogram vector; and

a creating code segment that causes the computer to create a second histogram vector that includes a vector element for each of the identified subsets that are included in the first histogram vector, with data within the vector elements of the second histogram representing the image features extracted.

21. The computer program of claim 15 wherein the weighting code segment includes:

an extracting code segment that causes the computer to extract one or more additional types of information about the image that differ from the image characteristics used to compute the joint histogram; and

a determining code segment that causes the computer to determine whether to modify a vector element value in the joint histogram based on the additional information that was extracted about the image, with the joint histogram including one or more vector elements.

22. The computer program of claim 15 wherein the applying code segment causes the computer to apply the weighting factor as the joint histogram is being computed.

23. The computer program of claim 15 wherein the determining code segment causes the computer to determine the weighting factor as the joint histogram is being computed.

24. The computer program of claim 15 wherein the determining code segment causes the computer to determine the weighting factor based on a centeredness image characteristic extracted about the image.

25. The computer program of claim 15 wherein the weighting factor is a non-zero weighting factor.

26. The computer program of claim 15 wherein the determining code segment causes the computer to determine the weighting factor based on at least one image characteristic that differs from the image characteristics used to compute the joint histogram.

27. A system for representing an image, comprising:

an extracting component that is structured and arranged to extract more than one image characteristic from an image;

a computing component that is structured and arranged to compute a joint histogram based on the more than one image characteristic extracted about the image;

a determining component that is structured and arranged to determine a weighting factor based on at least one of a position of a pixel within the image, color, brightness, edge density, texturedness, gradient magnitude, centeredness, and rank; and

an applying component that is structured and arranged to apply the weighting factor to at least one of the image characteristics extracted about the image by modifying a pixel count related to the image characteristic such that the joint histogram represents a weighted joint histogram.

28. The system of claim 27 further comprising a calculating component that is structured and arranged to calculate a posterized joint histogram.

29. The system of claim 27 further comprising a calculating component that is structured and arranged to calculate a posterized joint histogram after the weighting component applies the weighting factor.

30. The system of claim 27 wherein the extracting component is structured and arranged to extract information reflecting more than one feature of the image.

31. The system of claim 30 wherein:

the computing component is structured and arranged to compute a first histogram vector based on the information reflecting several features of the image, and

the first histogram vector includes one or more vector elements, each representing information for a different combination of features extracted about the image.

32. The system of claim 31 further comprising a calculating component that is structured and arranged to calculate a posterized joint histogram, wherein the calculating component is further structured and arranged to identify multiple subsets within at least one of the vector elements included in the first histogram vector, and to create a second histogram vector that includes a vector element for each of the identified subsets that are included in the first histogram vector, with data within the vector elements of the second histogram representing the image features extracted.

33. The system of claim 27 wherein:

the extracting component is structured and arranged to extract one or more additional types of information about the image that differ from the types of information used to compute the joint histogram and the determining component is structured and arranged to determine whether to modify a vector element value in the joint histogram based on the additional information that was extracted about the image, and

the joint histogram includes one or more vector elements.

34. The system of claim 27 wherein the applying component is structured and arranged to apply the weighting factor as the joint histogram is being computed.

35. The system of claim 27 wherein the determining component is structured and arranged to determine the weighting factor as the joint histogram is being computed.

36. The system of claim 27 wherein the determining component is structured and arranged to determine the weighting factor based on a centeredness image characteristic extracted about the image.

37. The system of claim 27 wherein the weighting factor is a non-zero weighting factor.

38. The system of claim 27 wherein the determining component is structured and arranged to determine the weighting factor based on at least one image characteristic that differs from the image characteristics used to compute the joint histogram.