Indexing of images and/or text

• US 6,522,780 B1
• Filed: 01/31/2001
• Issued: 02/18/2003
• 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): Tran, Phuoc
• Assistant Examiner(s): SHERALI, ISHRAT I
• Application Number:
  US09/772,879
• Time in Patent Office: 748 Days
• Field of Search: 382/218, 382/209, 382/219-220, 382/168, 382/170, 382/162, 382/165, 382/305, 382/108, 382/224, 358/403, 358/518, 348/522, 348/403.1, 345/700, 345/738, 345/835, 345/788, 345/419, 345/589, 707/1-6, 707/104.1, 707/520-522, 707/515-516
• US Class Current: 382/170
• CPC Class Codes: G06F 16/5838 using colour
• CPC Class Codes: Y10S 707/99936 Pattern matching access

Abstract

Indexing images for use in a searchable index includes receiving an image. A joint histogram may be computed for the received image such that the joint histogram includes joint features. A searchable index based on the joint features within the joint histogram may be created using an index methodology. Text also may be received and a searchable index of the received text may be created. A weighting factor may be applied to at least one of the joint features within the joint histogram such that the joint histogram represents a weighted joint histogram. A searchable index may include a first joint histogram feature corresponding to at least a first image, which includes the first joint histogram feature. The searchable index also may include a second joint histogram feature corresponding to at least a second image, which includes the second joint histogram feature.

Claims

1. A method for indexing images for use in a searchable index, the method comprising:

receiving an image;

computing a joint histogram for the image, wherein the joint histogram includes joint features; and

creating a searchable index based on the joint features within the joint histogram using an index methodology, wherein the searchable index includes individual joint features to which associated images correspond.

2. The method as in claim 1 wherein receiving the image includes receiving text and creating the searchable index includes creating the searchable index based on the received text.

3. The method as in claim 1 wherein computing the joint histogram includes applying a weighting factor to at least one of the joint features within the joint histogram such that the joint histogram represents a weighted joint histogram.

4. The method as in claim 1 wherein the searchable index is an inverted index such that creating the searchable index includes indexing the joint features within the joint histogram in an inverted indexing methodology.

5. The method as in claim 4 wherein the inverted index includes individual joint features to which associated images correspond.

6. The method as in claim 1 further comprising calculating a posterized joint histogram for the image by identifying subsets of the joint features of the joint histogram, wherein the posterized joint histogram includes a set of posterized joint features such that creating the searchable index includes creating the searchable index of the posterized joint features within the posterized joint histogram using an index methodology, and the searchable index includes individual posterized joint features to which associated images correspond.

7. The method as in claim 6 wherein:

computing the joint histogram includes applying a weighting factor to at least one joint feature within the joint histogram such that the joint histogram represents a weighted joint histogram, and

calculating the posterized joint histogram includes identifying subsets of weighted joint features.

8. The method as in claim 6 wherein the searchable index is an inverted index such that creating the searchable index includes indexing the posterized joint features within the posterized joint histogram in an inverted indexing methodology.

9. The method as in claim 8 wherein the inverted index includes individual posterized joint features to which associated images correspond.

10. A computer program for indexing images, the computer program being stored on a computer readable medium or a propagated signal comprising:

a receiving code segment that causes the computer to receive an image;

a computing code segment that causes the computer to compute a joint histogram for the image, wherein the joint histogram includes joint features; and

a creating code segment that causes the computer to create a searchable index based on the joint features within the joint histogram using an index methodology, wherein the searchable index includes individual joint features to which associated images correspond.

11. The computer program of claim 10 wherein the receiving code segment further causes the computer to receive text and the creating code segment further causes the computer to create the searchable index based on the received text.

12. The computer program of claim 10 wherein the computing code segment includes a weighting code segment that causes the computer to apply a weighting factor to at least one of the joint features within the joint histogram such that the joint histogram represents a weighted joint histogram.

13. The computer program of claim 10 wherein the index is an inverted index such that the creating code segment includes an indexing code segment that causes the computer to index the joint features within the joint histogram in an inverted indexing methodology.

14. The computer program of claim 13 wherein the inverted index includes individual joint features to which associated images correspond.

15. The computer program of claim 10 further comprising a calculating code segment that causes the computer to calculate a posterized joint histogram for the image by identifying subsets of the joint features of the joint histogram, wherein the posterized joint histogram includes a set of posterized joint features such that the creating code segment causes the computer to create the searchable index of the posterized joint features within the posterized joint histogram using an index methodology, and the searchable index includes individual posterized joint features to which associated images correspond.

16. The computer program of claim 15 wherein:

the computing code segment includes a weighting code segment that causes the computer to apply a weighting factor to at least one of the joint features within the joint histogram such that the joint histogram represents a weighted joint histogram, and

the calculating code segment causes the computer to calculate the posterized joint histogram by identifying subsets of weighted joint features.

17. The computer program of claim 15 wherein the index is an inverted index such that the creating code segment includes an indexing code segment that causes the computer to index the posterized joint features within the posterized joint histogram in an inverted indexing methodology.

18. The computer program of claim 17 wherein the inverted index includes individual posterized joint features to which associated images correspond.

19. A system for indexing images, the system comprising:

a receiving component that is structured and arranged to receive an image;

a computing component that is structured and arranged to compute a joint histogram for the image, wherein the joint histogram includes joint features; and

an indexing component that is structured an arranged to create a searchable index based on the joint features within the joint histogram using an index methodology, wherein the searchable index includes individual joint features to which associated images correspond.

20. The system of claim 19 wherein the receiving component is structured and arranged to receive text and the indexing component is structured and arranged to create the searchable index based on the received text.

21. The system of claim 19 wherein the computing component includes a weighting component that is structured and arranged to apply a weighting factor to at least one of the joint features within the joint histogram such that the joint histogram represents a weighted joint histogram.

22. The system of claim 19 wherein the index is an inverted index such that the indexing component indexes the joint features within the joint histogram in an inverted indexing methodology.

23. The system of claim 22 wherein the inverted index includes individual joint features to which associated images correspond.

24. The system of claim 19 further comprising a calculating component that is structured and arranged to calculate a posterized joint histogram for the image by identifying subsets of the joint features of the joint histogram, wherein the posterized joint histogram includes a set of posterized joint features such that the indexing component creates a searchable index of the posterized joint features within the posterized joint histogram using an index methodology, and the searchable index includes individual posterized joint features to which associated images correspond.

25. The system of claim 24 wherein:

the computing component includes a weighting component that is structured and arranged to apply a weighting factor to at least one of the joint features within the joint histogram such that the joint histogram represents a weighted joint histogram, and

the calculating component calculates the posterized joint histogram by identifying subsets of the weighted joint features.

26. The system of claim 24 wherein the index is an inverted index such that indexing component indexes the posterized joint features within the posterized joint histogram in an inverted indexing methodology.

27. The system of claim 26 wherein the inverted index includes individual posterized joint features to which associated images correspond.