Method and mechanism for analyzing the texture of a digital image

• US 20070098259A1
• Filed: 12/22/2005
• Published: 05/03/2007
• Est. Priority Date: 10/31/2005
• Status: Active Grant

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Shah, Shesha, Sathish, Sriram
• Granted Patent: US 7,583,839 B2
• Time in Patent Office: Days
• US Class Current: 382/165
• CPC Class Codes: G06F 16/5862 using texture
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/90 Determination of colour characteristics
• CPC Class Codes: G06V 10/56 relating to colour

Abstract

A method and apparatus for comparing content attributes, e.g., texture, of one digital image a second digital image. An image operation is performed on gray scale image data to identify intensity transitions of a first digital image. The gray scale image data comprises a set of data values that each identifies a gray scale value. For at least a subset of the set of data values, a slope value for an intensity transition associated with the data value is determined. The slope values are converted to possible slope values of a bounded set of possible slope values. First slope data that describes a frequency of occurrence, across the subset of data values, of each possible slope value is generated. The first slope data is compared to second slope data associated with the second digital image to determine how similar the first digital image is to the second digital image.

Claims

1. A method for analyzing digital images, comprising:

performing an image operation on gray scale image data to identify intensity transitions of a first digital image,

wherein the gray scale image data is a digital representation of a gray scale version of the first digital image, and

wherein the gray scale image data comprises a set of data values, wherein each data value of the set of data values identifies a gray scale value of a sequence of gray scale values;

determining, for at least a subset of data values, in the set of data values, a slope value for an intensity transition associated with the data value;

converting the slope value, associated with each data value in the subset of data values, to a possible slope value of a bounded set of possible slope values; and

generating first slope data, for the first digital image, that describes a frequency of occurrence, across the subset of data values, of each possible slope value, of the bounded set of possible slope values.

2. The method of claim 1, further comprising:

comparing the first slope data to second slope data associated with a second digital image to determine how similar the first digital image is to the second digital image.

3. The method of claim 2, wherein comparing the first slope data to second slope data comprises determining the Euclidean distance between the first slope data and the second slope data.

4. The method of claim 1, further comprising:

generating the gray scale image data based on color image data, wherein the color image data is a digital representation of a color version of the first digital image.

5. The method of claim 1, wherein the image operation is one of:

a Laplacian of the Gaussian image operation, a Sobel image operation, and a Canny image operation.

6. The method of claim 1, wherein generating the first slope data comprises:

generating a histogram that describes a frequency of each possible slope value, in the bounded set of possible slope values, in the subset of data values;

normalizing the histogram by expressing the frequency of each possible slope value as a percentage of a total number of data values in the subset of data values.

7. The method of claim 1, further comprising:

analyzing the first slope data to determine whether the first digital image depicts either a certain type of terrain or a certain type of crop.

8. The method of claim 1, further comprising:

analyzing the first slope data to determine whether the first digital image depicts human skin.

9. The method of claim 1, further comprising:

analyzing the first slope data to determine whether the first digital image depicts offensive content.

10. A machine-readable medium carrying one or more sequences of instructions for analyzing digital images, wherein execution of the one or more sequences of instructions by one or more processors cause:

performing an image operation on gray scale image data to identify intensity transitions of a first digital image,

wherein the gray scale image data is a digital representation of a gray scale version of the first digital image, and

wherein the gray scale image data comprises a set of data values, wherein each data value of the set of data values identifies a gray scale value of a sequence of gray scale values;

determining, for at least a subset of data values, in the set of data values, a slope value for an intensity transition associated with the data value;

converting the slope value, associated with each data value in the subset of data values, to a possible slope value of a bounded set of possible slope values; and

generating first slope data, for the first digital image, that describes a frequency of occurrence, across the subset of data values, of each possible slope value, of the bounded set of possible slope values.

11. The machine-readable medium of claim 10, wherein execution of the one or more sequences of instructions by the one or more processors further cause:

comparing the first slope data to second slope data associated with a second digital image to determine how similar the first digital image is to the second digital image.

12. The machine-readable medium of claim 11, wherein comparing the first slope data to second slope data comprises determining the Euclidean distance between the first slope data and the second slope data.

13. The machine-readable medium of claim 10, wherein execution of the one or more sequences of instructions by the one or more processors further cause:

generating the gray scale image data based on color image data, wherein the color image data is a digital representation of a color version of the first digital image.

14. The machine-readable medium of claim 10, wherein the image operation is one of:

a Laplacian of the Gaussian image operation, a Sobel image operation, and a Canny image operation.

15. The machine-readable medium of claim 10, wherein generating the first slope data comprises:

generating a histogram that describes a frequency of each possible slope value, in the bounded set of possible slope values, in the subset of data values;

normalizing the histogram by expressing the frequency of each possible slope value as a percentage of a total number of data values in the subset of data values.

16. The machine-readable medium of claim 10, wherein execution of the one or more sequences of instructions by the one or more processors further cause:

analyzing the first slope data to determine whether the first digital image depicts either a certain type of terrain or a certain type of crop.

17. The machine-readable medium of claim 10, wherein execution of the one or more sequences of instructions by the one or more processors further cause:

analyzing the first slope data to determine whether the first digital image depicts human skin.

18. The machine-readable medium of claim 10, wherein execution of the one or more sequences of instructions by the one or more processors further cause:

analyzing the first slope data to determine whether the first digital image depicts offensive content.

19. An apparatus for analyzing digital images, comprising:

a machine-readable medium carrying one or more sequences of instructions; and

one or more processors, wherein execution of the one or more sequences of instructions by the one or more processors cause;

performing an image operation on gray scale image data to identify intensity transitions of a first digital image,

wherein the gray scale image data is a digital representation of a gray scale version of the first digital image, and

wherein the gray scale image data comprises a set of data values, wherein each data value of the set of data values identifies a gray scale value of a sequence of gray scale values;

determining, for at least a subset of data values, in the set of data values, a slope value for an intensity transition associated with the data value;

converting the slope value, associated with each data value in the subset of data values, to a possible slope value of a bounded set of possible slope values; and

generating first slope data, for the first digital image, that describes a frequency of occurrence, across the subset of data values, of each possible slope value, of the bounded set of possible slope values.

20. The apparatus of claim 19, wherein execution of the one or more sequences of instructions by the one or more processors further cause:

comparing the first slope data to second slope data associated with a second digital image to determine how similar the first digital image is to the second digital image.

21. The apparatus of claim 20, wherein comparing the first slope data to second slope data comprises determining the Euclidean distance between the first slope data and the second slope data.

22. The apparatus of claim 19, wherein execution of the one or more sequences of instructions by the one or more processors further cause:

generating the gray scale image data based on color image data, wherein the color image data is a digital representation of a color version of the first digital image.

23. The apparatus of claim 19, wherein the image operation is one of:

a Laplacian of the Gaussian image operation, a Sobel image operation, and a Canny image operation.

24. The apparatus of claim 19, wherein generating the first slope data comprises:

generating a histogram that describes a frequency of each possible slope value, in the bounded set of possible slope values, in the subset of data values;

normalizing the histogram by expressing the frequency of each possible slope value as a percentage of a total number of data values in the subset of data values.

25. The apparatus of claim 19, wherein execution of the one or more sequences of instructions by the one or more processors further cause:

analyzing the first slope data to determine whether the first digital image depicts either a certain type of terrain or a certain type of crop.

26. The apparatus of claim 19, wherein execution of the one or more sequences of instructions by the one or more processors further cause:

analyzing the first slope data to determine whether the first digital image depicts human skin.

27. The apparatus of claim 19, wherein execution of the one or more sequences of instructions by the one or more processors further cause:

analyzing the first slope data to determine whether the first digital image depicts offensive content.