Method for automatically comparing content of images for classification into events

• US 6,351,556 B1
• Filed: 11/20/1998
• Issued: 02/26/2002
• Est. Priority Date: 11/20/1998
• Status: Expired due to Term

• Current Assignee: Monument Peak Ventures, LLC (Dominion Harbor Enterprises, LLC)
• Original Assignee: Eastman Kodak Company
• Inventors: Loui, Alexander C., Pavie, Eric S.
• Primary Examiner(s): Tran, Phuoc
• Assistant Examiner(s): Alavi, Amir
• Application Number:
  US09/197,363
• Time in Patent Office: 1,194 Days
• Field of Search: 382/162, 382/163, 382/164, 382/165, 382/166, 382/167, 382/168, 382/169, 382/170, 382/171, 382/172, 382/305
• US Class Current: 382/164
• CPC Class Codes: G06T 7/00 Image analysis
• CPC Class Codes: G06V 10/507 Summing image-intensity values; Histogram projection analysis
• CPC Class Codes: G06V 10/758 Involving statistics of pixels or of feature values, e.g. histogram matching

Abstract

A method for comparing image content of first and second images, the method comprises the steps of extracting a portion of both the first and second images both of which portions are determined to include a main subject area of each image; dividing the main subject area of the images into a plurality of blocks; computing a color histogram for one block in each image; computing a histogram intersection value between the block of the first image and the block of the second image; and determining a first threshold value for the computed histogram intersection value that determines similarity between the block in the first image and the block in the second image.

Claims

1. A method for comparing image content of first and second images, the method comprising the steps of:

(a) extracting a portion of both the first and second images both of which portions are determined to include a main subject area of each image;

(b) dividing the main subject area of the images into a plurality of blocks;

(c) computing a color histogram for one block in each image;

(d) computing a histogram intersection value between the block of the first image and the block of the second image;

(e) determining a first threshold value for the computed histogram intersection value that determines similarity between the block in the first image and the block in the second image; and

(f) shifting the first image in a predetermined direction if there is not a similarity between the first and second images and redetermining the similarity between the images in step (c)-(e) for blocks that are shifted with respect to each other.

2. The method as in claim 1, wherein step (c) includes (c1) computing a color histogram for each of the plurality of blocks, and step (d) includes (d1) computing a histogram intersection value between one block in the first image and a plurality of blocks in the second image.

3. The method as in claim 2 further comprising repeating step (d1) if there is not similarity between the first and second images.

4. The method as in claim 1, wherein step (a) includes dividing both of the images into three portions and using a middle portion of the three portions as the main subject area.

5. The method as in claim 1 further comprising the step of determining a second threshold value for the computed histogram intersection value that determines similarity between the block in the first image and the block in the second image and consequently if the two images are near-duplicates of each other.

6. A computer program product for comparing image content of first and second images comprising:

a computer readable storage medium having a computer program stored thereon for performing the steps of;

(a) extracting a portion of both the first and second images both of which portions are determined to include a main subject of each image;

(b) dividing the main subject of the images into a plurality of blocks;

(c) computing a color histogram for one block in each image;

(d) computing a histogram intersection value between the block of the first image and the block of the second image;

(e) determining a threshold value for the computed histogram intersection value that determines similarity between the block in the first image and the block in the second image; and

(f) shifting the first image in a predetermined direction if there is not a similarity between the first and second images and redetermining the similarity between the images in step (c)-(e) for blocks that are shifted with respect to each other.

7. The computer program product as in claim 6, wherein step (c) includes (c1) computing a color histogram for each of the plurality of blocks, and step (d) includes (d1) computing a histogram intersection value between one block in the first image and a plurality of blocks in the second image.

8. The computer program product as in claim 7 further comprising repeating step (d1) if there is not similarity between the fist and second images.

9. The computer program product as in claim 8, wherein step (a) includes dividing both of the images into three portions and using a middle portion of the three portions as the main subject area.

10. The computer program product as in claim 6 further comprising the step of determining a second threshold value for the computed histogram intersection value that determines similarity between the block in the first image and the block in the second image and consequently if the two images are near-duplicates of each other.

11. A method for comparing image content of first and second images, the method comprising the steps of:

(a) extracting a portion of both the first and second images both of which portions are determined to include a main subject area of each image;

(b) dividing the main subject area of the images into a plurality of blocks;

(c) computing a color histogram for each of the plurality of blocks in each image;

(d) computing a histogram intersection value between one block in the first image and two or more blocks in the second image; and

,

(e) determining a first threshold value for the computed histogram intersection value that determines similarity between the block in the first image and the blocks in the second image.

12. The method as in claim 11 further comprising shifting the first image in a predetermined direction if there is not similarity between the first and second images.

13. The method as in claim 11, wherein step (a) includes dividing both of the images into three portions and using a middle portion of the three portions as the mane subject area.

14. The method as in claim 11 further comprising the step of determining a second threshold value for the computed histogram intersection value that determines similarity between the block in the first image and the blocks in the second image and consequently if the two images are near-duplicates of each other.

15. A computer storage medium having instructions stored therein for causing a computer to perform the method of claim 11.

16. A method for comparing image content of first and second images, the method comprising the steps of:

(a) dividing the first and second images into three portions and using a middle portion of the three portions as a main subject area of each image;

(b) dividing the main subject area of the images into a plurality of blocks;

(c) computing a color histogram for one block in each image;

(d) computing a histogram intersection value between the block in the first image and the block in the second image; and

(e) determining a first threshold value for the computed histogram intersection value that determines similarity between the block in the first image and the block in the second image.

17. The method as in claim 16 further comprising the step of determining a second threshold value for the computed histogram intersection value that determines similarity between the block in the first image and the block in the second image and consequently if the two images are near-duplicates of each other.

18. A computer storage medium having instructions stored therein for causing a computer to perform the method of claim 16.