Identifying image type in a capture system

• US 7,657,104 B2
• Filed: 11/21/2005
• Issued: 02/02/2010
• Est. Priority Date: 11/21/2005
• Status: Expired due to Fees

• Current Assignee: McAfee, LLC
• Original Assignee: McAfee, Inc. (McAfee, LLC)
• Inventors: Deninger, William, de la Iglesia, Erik
• Primary Examiner(s): Do; Anh Hong
• Application Number:
  US11/284,553
• Publication Number:
  US 20070116366A1
• Time in Patent Office: 1,534 Days
• Field of Search: 382/224, 382/181, 382/233, 382/165, 382/305, 382/232, 382/173, 382/240, 382/261, 382/268, 382/199, 707/104.1, 358/403, 358/404, 358/444, 705/3, 348/E5.076, 348/E5.077
• US Class Current: 382/224
• CPC Class Codes: G06F 21/55 Detecting local intrusion or implementing counter-measures
• 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: H04L 63/123 received data contents, e.g. message integrity

Abstract

Visual objects can be classified according to image type. In one embodiment, the present invention includes capturing a visual object, and decompressing the visual object to a colorspace representation exposing each pixel. The contribution of each pixel to a plurality of image types can then be determined. Then, the contributions can be combined, and the image type of the visual object can be determined based on the contributions.

Claims

1. A method to be executed by a processor in an electronic environment, comprising:

capturing a visual object being transmitted over a network;

selecting a subset of the visual object;

decompressing the selected subset of the visual object to a colorspace representation to expose each pixel of the selected subset of the visual object;

determining a contribution of each pixel to a plurality of image types;

determining an image type of the visual object based on the contribution of each pixel to the plurality of image types, the determined image type being one of the plurality of image types.

2. The method of claim 1, wherein selecting the subset of the visual object comprises selecting the entire visual object.

3. The method of claim 1, wherein selecting the subset of the visual object comprises selecting the subset of the visual object based on the size and dimensions of the visual object.

4. The method of claim 1, wherein determining the contribution of each pixel to a plurality of image types comprises mapping a pixel to a plurality of values, each value of the plurality of values quantifying a contribution of the pixel to an image type of the plurality of image types.

5. The method of claim 4, wherein the mapping of the pixel to the plurality of values is based on a position of the pixel within the visual object.

6. The method of claim 4, wherein mapping the pixel to the plurality of values comprises mapping the pixel to a plurality of values between 0 and 2, wherein 0 indicates no contribution, 1 indicates a weak contribution, and 2 indicates a strong contribution.

7. The method of claim 1, wherein determining the image type of the visual object comprises classifying the visual object as belonging to the image type if the sum of the contributions of each pixel to the image type exceeds a threshold.

8. The method of claim 1, wherein the plurality of image types comprise skintone, x-ray, greyscale, graph, astrological, geographic, circuit design mask, banner advertisement.

9. A capture device comprising:

one or more capture modules to capture a visual object being transmitted over a network;

a scaling module to select a subset of the visual object;

a decompressor to decompress the selected subset of the visual object to a colorspace representation thereby exposing each pixel of the selected subset of the visual object;

a pixel map module to determine a contribution of each pixel to a plurality of image types; and

a classification module to determine an image type of the visual object based on the contribution of each pixel to the plurality of image types, the determined image type being one of the plurality of image types.

10. The capture device of claim 9, wherein the scaling module selects the subset of the visual object by selecting the entire visual objects.

11. The capture device of claim 9, wherein the scaling module selects the subset of the visual object by selecting the subset of the visual object based on the size and dimensions of the visual object.

12. The capture device of claim 9, wherein the pixel map module determines the contribution of each pixel to a plurality of image types by mapping a pixel to a plurality of values, each value of the plurality of values quantifying a contribution of the pixel to an image type of the plurality of image types.

13. The capture device of claim 12, wherein mapping the pixel to the plurality of values comprises accessing a pixel mapping table associating the pixel with a plurality of contribution mappings.

14. The capture device of claim 13, wherein the pixel map module selects one of the plurality of contribution mappings based on a position of the pixel within the visual object.

15. The capture device of claim 9, wherein the classification module determines the image type of the visual object by classifying the visual object as belonging to the image type if the sum of the contributions of each pixel to the image type exceeds a threshold.

16. A machine-readable medium having stored thereon data representing instructions that, when executed by a processor, cause the processor to perform operations comprising:

capturing a visual object being transmitted over a network;

selecting a subset of the visual object;

decompressing the selected subset of the visual object to a colorspace representation to expose each pixel of the selected subset of the visual object;

determining a contribution of each pixel to a plurality of image types;

determining an image type of the visual object based on the contribution of each pixel to the plurality of image types, the determined image type being one of the plurality of image types.

17. The machine-readable medium of claim 16, wherein selecting the subset of the visual object comprises selecting the entire visual object.

18. The machine-readable medium of claim 16, wherein selecting the subset of the visual object comprises selecting the subset of the visual object based on the size and dimensions of the visual object.

19. The machine-readable medium of claim 16, wherein determining the contribution of each pixel to a plurality of image types comprises mapping a pixel to a plurality of values, each value of the plurality of values quantifying a contribution of the pixel to an image type of the plurality of image types.

20. The machine-readable medium of claim 19, wherein the mapping of the pixel to the plurality of values is based on a position of the pixel within the visual object.

21. The machine-readable medium of claim 19, wherein mapping the pixel to the plurality of values comprises mapping the pixel to a plurality of values between 0 and 2, wherein 0 indicates no contribution, 1 indicates a weak contribution, and 2 indicates a strong contribution.

22. The machine-readable medium of claim 16, wherein determining the image type of the visual object comprises classifying the visual object as belonging to the image type if the sum of the contributions of each pixel to the image type exceeds a threshold.