METHOD, APPARATUS AND STORAGE MEDIUM FOR IMAGE CLASSIFICATION

• US 20090022407A1
• Filed: 02/08/2007
• Published: 01/22/2009
• Est. Priority Date: 02/09/2006
• Status: Active Grant

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Information Technology Services Incorporated (Canon Inc.)
• Inventors: Ji, Xin, Wang, Jianmin
• Granted Patent: US 8,238,670 B2
• Time in Patent Office: Days
• US Class Current: 382/224
• CPC Class Codes: G06V 10/50 by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
• CPC Class Codes: G06V 10/56 relating to colour
• CPC Class Codes: G06V 20/10 Terrestrial scenes scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59

Abstract

In an image classification method, dividing an input image into blocks; obtaining block features of each block of the image; performing an evaluation of each block based on the block features thereof; obtaining image features based on the evaluations of the blocks of the image; and classifying the image based on the image features into pre-defined categories.

Claims

1. An image classification method, comprising the steps of:

dividing an input image into blocks;

obtaining block features of each block of the image;

performing an evaluation of each block based on the block features thereof;

obtaining image features based on the evaluations of the blocks of the image; and

classifying the image based on the image features into pre-defined categories.

2. An image classification method according to claim 1, further comprising the step of resizing the input image to predetermined size before dividing.

3. An image classification method according to claim 2, wherein, in the step of resizing, the ratio of image width and height is not changed.

4. An image classification method according to claim 1, wherein the block features of a block are color statistic values of the pixels in the block.

5. An image classification method according to claim 4, wherein the color statistic values are mean values, mean square variances or covariance of the values.

6. An image classification method according to claim 1, wherein the evaluation of each block is obtained by summing up the respectively weighted block features of the block.

7. An image classification method according to claim 1, wherein the image features of the image are statistic values of the evaluations of the blocks of the image.

8. An image classification method according to claim 7, wherein the statistic values are mean values or mean square variances.

9. An image classification method according to claim 8, wherein the mean values or mean square variances are that of some largest evaluations of the blocks of the image.

10. An image classification method according to claim 1, wherein, in the classifying step, a classification value is obtained by summing up the respectively weighted image features of the image.

11. An image classification method according to claim 10, wherein, in the classifying step, comparing the classification value with a predetermined value to determine whether the image is classified into a category of a pre-defined content or not.

12. An image classification method according to claim 11, wherein the classification value indicates the degree to which the image contains the pre-defined content.

13. An image classification method according to claim 1, wherein the evaluation of a block indicates the degree to which the block contains the pre-defined content.

27. A computer readable program stored in a storage medium for classifying an image, the program including instructions for causing a computer to implement the method according to any one of claims 1 to 13.

14. An image classification apparatus, comprising:

image dividing means for dividing an input image into blocks;

block feature obtaining means for obtaining the block features of each block of the image;

block evaluating means for performing an evaluation of each block based on the block features thereof;

image feature obtaining means for obtaining image features based on the evaluations of the blocks of the image; and

image classification means for classifying the image based on the image features into pre-defined categories.

15. An image classification apparatus according to claim 14, further comprising image resizing means for resizing the input image to a predetermined size before being divided by the image dividing means.

16. An image classification apparatus according to claim 15, wherein when the image resizing means resizes the input image, the image resizing means does not change the ratio of image width and height.

17. An image classification apparatus according to claim 14, wherein the block features of a block are color statistic values of the pixels in the block.

18. An image classification apparatus according to claim 17, wherein the color statistic values are mean values, mean square variances or covariance of the values.

19. An image classification apparatus according to claim 14, wherein the block evaluating means obtains the evaluation of each block by summing up the respectively weighted block features of the block.

20. An image classification apparatus according to claim 14, wherein the image features of the image are statistic values of the evaluations of the blocks of the image.

21. An image classification apparatus according to claim 20, wherein the statistic values are mean values or mean square variances.

22. An image classification apparatus according to claim 20, wherein the mean values or mean square variances are that of some largest evaluations of the blocks of the image.

23. An image classification apparatus according to claim 14, wherein the image classification means obtains a classification value of the image by summing up the respectively weighted image features of the image.

24. An image classification apparatus according to claim 23, wherein the image classification means compares the classification value with a predetermined value to determine whether the image is classified into a category of a pre-defined content or not.

25. An image classification apparatus according to claim 24, wherein the classification value indicates the degree to which the image contains the pre-defined content.

26. An image classification apparatus according to claim 14, wherein the evaluation of a block indicates the degree to which the block contains the pre-defined content.