System and method for effectively implementing a texture feature detector

• US 20070031035A1
• Filed: 08/04/2005
• Published: 02/08/2007
• Est. Priority Date: 08/04/2005
• Status: Active Grant

• Current Assignee: Sony Corporation (Sony Group Corp.), Sony Electronics Inc. (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.), Sony Electronics Inc. (Sony Group Corp.)
• Inventors: Hovden, Gunnar
• Granted Patent: US 7,609,885 B2
• Time in Patent Office: Days
• US Class Current: 382/170
• CPC Class Codes: G06T 7/40 Analysis of texture depth or shape recovery from texture G06T7/529

Abstract

A system and method are disclosed for performing an image search procedure in which a histogram calculator initially performs image analysis procedures to construct horizontal and vertical texture histograms corresponding to texture characteristics of subject images. A feature detector may then perform the image search procedure by comparing the texture histograms to identify one or more matching images in which texture characteristics are most similar.

Claims

1. A system for performing an image search procedure with an electronic device, comprising:

a histogram calculator that performs image analysis procedures to construct texture histograms corresponding to texture characteristics of images accessed by said electronic device; and

a feature detector that performs said image search procedure by comparing said texture histograms to identify one or more matching images in which said texture characteristics are similar.

2. The system of claim 1 wherein said texture histograms include a horizontal texture histogram and a vertical texture histogram for each of said images.

3. The system of claim 1 wherein said histogram calculator initially converts said images into a grey-scale format.

4. The system of claim 1 wherein said histogram calculator divides said images into sets of pixel blocks.

5. The system of claim 4 wherein each of said pixel blocks has pixel dimensions of sixteen horizontal pixels by sixteen vertical pixels.

6. The system of claim 4 wherein said histogram calculator performs Discrete Fourier Transform procedures upon each of said pixel blocks to produce DFT coefficients.

7. The system of claim 6 wherein said DFT coefficients correspond to horizontal pixel rows and vertical pixel columns from said each of said pixel blocks.

8. The system of claim 7 wherein said histogram calculator sums said DFT coefficients for said horizontal pixel rows to produce summed horizontal DFT coefficients for said each of said pixel blocks, said histogram calculator also summing said DFT coefficients for said vertical pixel columns to produce summed vertical DFT coefficients for said each of said pixel blocks.

9. The system of claim 6 wherein said histogram calculator disregards a lowest DC coefficient from said DC coefficients.

10. The system of claim 6 wherein said histogram calculator disregards upper DC coefficients eight through sixteen from said DC coefficients.

11. The system of claim 6 wherein said histogram calculator identifies horizontal peak values corresponding to maximum horizontal DFT coefficients from horizontal pixel rows of said pixel blocks.

12. The system of claim 11 wherein said histogram calculator populates horizontal texture histograms for said images by utilizing said horizontal peak values.

13. The system of claim 6 wherein said histogram calculator identifies vertical peak values corresponding to maximum vertical DFT coefficients from vertical pixel columns of said pixel blocks.

14. The system of claim 13 wherein said histogram calculator populates vertical texture histograms for said images by utilizing said vertical peak values.

15. The system of claim 6 wherein said histogram calculator sets peak values for populating said texture histograms equal to corresponding DC coefficients if said peak values are less than a predetermined threshold value.

16. The system of claim 1 wherein said feature detector compares horizontal texture histograms for each of said images during said image search procedure to determine a horizontal texture similarity score, said feature detector also comparing vertical texture histograms for each of said images during said image search procedure to determine a vertical texture similarity score.

17. The system of claim 16 wherein said feature detector calculates an overall similarity score for each pair of said images by taking the worst of either said horizontal texture similarity score or said vertical texture similarity score.

18. The system of claim 1 wherein said feature detector compares said texture histograms by calculating an Earth Mover'"'"'s Distance value for pairs of said texture histograms corresponding to said images.

19. The system of claim 1 wherein said texture histograms include a horizontal texture histogram, a vertical texture histogram, and a diagonal texture histogram for each of said images.

20. A method for performing an image search procedure with an electronic device, comprising the steps of:

performing image analysis procedures with a histogram calculator that constructs texture histograms corresponding to texture characteristics of images accessed by said electronic device; and

utilizing a feature detector to perform said image search procedure by comparing said texture histograms to identify one or more matching images in which said texture characteristics are similar.