Granted Patent B2

US 9,305,208 · App. 13/740,136 · Granted Apr 5, 2016

Abstract

Claims (54)

1. A method for categorizing at least one image, the method comprising:

obtaining, by an image procurement interface of a computing apparatus, the at least one image, the at least one image containing a plurality of pixel blocks, each of the pixel blocks containing a plurality of pixels from the at least one image;

for each of the pixel blocks, mapping, by a grid generation module of the computing apparatus, the pixel block to a cell of a first two-dimensional grid, wherein each of the cells identifies whether a corresponding pixel block from the at least one image contains more than a threshold number of skin-colored pixels;

classifying, by a grid classification module of the computing apparatus, the first grid, wherein classifying the first grid includes determining whether the first grid is indicative of an offensive characteristic by comparing locations of the cells of the first two-dimensional grid which identify pixel blocks that contain more than the threshold number of skin-colored pixels to a reference two-dimensional Boolean grid known to represent the offensive characteristic;

if the first grid is determined to be indicative of the offensive characteristic, categorizing, by a categorization module of the computing apparatus, the at least one image as offensive; and

if the first grid is determined to be not indicative of the offensive characteristic, categorizing, by the categorization module of the computing apparatus, the at least one image as not offensive.

2. The method of claim 1 , wherein the offensive characteristic is an offensive pose.

3. The method of claim 2 , wherein the offensive pose is a pornographic pose.

4. The method of claim 1 , wherein categorizing the at least one image as offensive comprises categorizing the at least one image as a pornographic image.

5. The method of claim 1 further comprising:

processing the at least one image by transforming the at least one image into a hue saturation brightness (HSB) color space, classifying hue components and saturation components associated with the at least one image, and ignoring brightness components associated with the at least one image.

6. The method of claim 1 , further comprising:

determining whether a first one of the pixels blocks contains pixels that are predominantly in a first range of colors; and

if the first pixel block contains pixels that are predominantly in the first range of colors, marking a first one of the cells to indicate that the first pixel block contains pixels that are predominantly in the first range of colors.

7. The method of claim 6 , wherein the first range of colors is a range of skin colors.

8. The method of claim 1 , wherein the first grid is a two-dimensional Boolean grid.

9. The method of claim 1 , wherein obtaining the at least one image includes obtaining a first image and a second image, the first image and the second image being part of one media presentation, and wherein mapping the at least one image to at least the first grid comprises mapping the first image to the first grid and mapping the second image to a second grid, the method further comprising:

classifying the second grid, wherein classifying the second grid includes determining whether the second grid is indicative of the offensive characteristic;

if the second grid is determined to be indicative of the offensive characteristic, categorizing the second image as offensive;

if the second grid is determined to be not indicative of the offensive characteristic, categorizing the second image as not offensive; and

if the first image and the second image are categorized to be offensive, categorizing the media presentation as offensive.

10. The method of claim 1 , wherein the at least one image is a thumbnail image.

11. A non-transitory computer-readable medium comprising computer program code, the computer program code that, when executed by a processor, causes the processor to:

obtain the at least one image, the at least one image containing a plurality of pixel blocks, each of the pixel blocks containing a plurality of pixels from the at least one image;

for each of the pixel blocks, map the pixel block to a cell of a first two-dimensional grid, wherein each of the cells identifies whether a corresponding pixel block from the at least one image contains more than a threshold number of skin-colored pixels;

classify the first grid, wherein classifying the first grid includes determining whether the first grid is indicative of an offensive characteristic by comparing locations of the cells of the first two-dimensional grid which identify pixel blocks that contain more than the threshold number of skin-colored pixels to a reference two-dimensional Boolean grid known to represent the offensive characteristic;

if the first grid is determined to be indicative of the offensive characteristic, categorize the at least one image as offensive; and

if the first grid is determined to be not indicative of the offensive characteristic, categorize the at least one image as not offensive.

12. The non-transitory computer-readable medium of claim 11 , wherein the offensive characteristic is a pornographic pose.

13. The non-transitory computer-readable medium of claim 12 wherein categorizing the at least one image as offensive comprises categorizing the at least one image as a pornographic image.

14. The non-transitory computer-readable medium of claim 11 , further comprising computer program code that, when executed by the processor, causes the processor to:

process the at least one image by transforming the at least one image into a hue saturation brightness (HSB) color space, classifying hue components and saturation components associated with the at least one image, and ignoring brightness components associated with the at least one image.

15. The non-transitory computer-readable medium of claim 11 , further comprising computer program code that, when executed by the processor, causes the processor to:

determine whether a first one of the pixels blocks contains pixels that are predominantly in a first range of colors; and

if the first pixel block contains pixels that are predominantly in the first range of colors, mark a first one of the cells to indicate that the first pixel block contains pixels that are predominantly in the first range of colors.

16. The non-transitory computer-readable medium of claim 15 , wherein the first range of colors is a range of skin colors.

17. The non-transitory computer-readable medium of claim 11 , wherein the first grid is a two-dimensional Boolean grid.

18. The non-transitory computer-readable medium of claim 11 , wherein obtaining the at least one image comprises obtaining a first image and a second image, the first image and the second image being part of one media presentation, and wherein mapping the at least one image to at least the first grid comprises mapping the first image to the first grid and mapping the second image to a second grid, the computer-readable medium further comprises computer program code that, when executed by the processor, causes the processor to:

classify the second grid, wherein classifying the second grid includes determining whether the second grid is indicative of the offensive characteristic;

if the second grid is determined to be indicative of the offensive characteristic, categorize the second image as offensive;

if the second grid is determined to be not indicative of the offensive characteristic, categorize the second image as not offensive; and

if the first image and the second image are categorized to be offensive, categorizing the media presentation as offensive.

19. The non-transitory computer-readable medium of claim 11 , wherein the at least one image is a thumbnail image.

20. An apparatus for categorizing at least one image, the apparatus comprising:

a memory;

a processor communicatively coupled to the memory; and

software instructions stored in the memory that, when executed by the processor, cause the processor to:

obtain the at least one image, the at least one image containing a plurality of pixel blocks, each of the pixel blocks containing a plurality of pixels from the at least one image;

for each of the pixel blocks, map the pixel block to a cell of a two-dimensional grid, wherein each of the cells identifies whether a corresponding pixel block from the at least one image contains more than a threshold number of skin-colored pixels;

classify the grid, wherein classifying the grid includes determining whether the grid is indicative of an offensive characteristic by comparing locations of the cells of the two-dimensional grid which identify pixel blocks that contain more than the threshold number of skin-colored pixels to a reference two-dimensional Boolean grid known to represent the offensive characteristic;

if the grid is determined to be indicative of the offensive characteristic, categorize the at least one image as offensive; and

if the grid is determined to be not indicative of the offensive characteristic, categorize the at least one image as not offensive.

21. The apparatus of claim 20 , wherein the memory further stores software instructions that, when executed by the processor, cause the processor to identify a set of pixel blocks included in the plurality of pixel blocks which contain pixels that are predominantly in a first range of colors.

22. The apparatus of claim 21 , wherein the first range of colors is a range of skin colors.

Assignments (9)