SPAM DETECTION WITHIN IMAGES OF A COMMUNICATION

• US 20090100523A1
• Filed: 04/30/2004
• Published: 04/16/2009
• Est. Priority Date: 04/30/2004
• Status: Abandoned Application

• Current Assignee: Harris Technology LLC
• Original Assignee: Harris Technology LLC
• Inventors: Harris, Scott C.
• Application Number:
  US10/835,111
• Publication Number:
  US 20090100523A1
• Time in Patent Office: Days
• US Class Current: 726/26
• CPC Class Codes: G06V 20/62 Text, e.g. of license plates, overlay texts or captions on TV images
• CPC Class Codes: H04L 51/212 using filtering or selective blocking

Abstract

Determining undesirable, or “spam” communication, by reviewing and recognizing portions within the communications that are things other than ASCII or text. Images are analyzed to determine whether the content of the images is likely to represent undesired content. The images can be classified as to type, can be OCRed, and the contents of the recognition used for analysis, and can be compared against similar images in a database.

Claims

1. A method comprising:

determining non-text parts in an electronic communication; and

analyzing said non text parts, to determine information in said non-text part which indicate that the electronic communication is an undesired communication.

2. A method as in claim 1, wherein said analyzing comprises analyzing an image as said non text part.

3. A method as in claim 2, wherein said analyzing comprises optically character recognizing words in said non text part, and analyzing said words to determine an undesired communication.

4. A method as in claim 3, further comprising analyzing text parts in the communication using a heuristic engine and wherein said analyzing said words comprises heuristic analysis of said words in said non-text part using the same heuristic engine.

5. A method as in claim 2, wherein said analyzing comprises automatically determining a category of the image by comparing the image with a catalog of image information that includes known image information therein, where said automatically determining determines multiple said categories, where at least one of the known information represents an undesired category, and determining if the category represents said undesired category.

6. A method as in claim 2, wherein said analyzing comprises determining a hash of at least portions of said image and comparing said hash of said portions of the image against other hashes of other at least portions of other images known to represent undesired content.

7. A method as in claim 5, wherein said comparing determines multiple different undesirable categories.

8. A system, comprising:

a communication device, which receives an electronic communication from a channel; and

a processing part, which processes said electronic communication, and analyzes a non-text part of the communication, to determine undesired communications.

9. A system as in claim 8, wherein said processing part includes a computer, which is programmed for said processing.

10. A system as in claim 8, wherein said processing part analyzes an image as said non text part.

11. A system as in claim 10, wherein said analyzes comprises optically character recognizing text within the image, and analyzing the optically character recognized text to determine that the communication is undesirable.

12. A system as in claim 10, wherein said analyzes comprises using the processing part to automatically categorize the image by comparing the image with a catalog of image information that includes known image information therein, where said automatically determining determines multiple said categories, where at least one of the known information represents an undesired category, and to use a category of the image to determine that the communication is undesirable.

13. A system as in claim 10, further comprising a database of image parts, at least some of said image parts representing images from known undesirable communications, wherein said analyzes comprises using the processing part to automatically compare the image to image parts in said database.

14. A system as in claim 8, wherein said processing part further includes a heuristic engine analyzing text parts in the communication and also analyzes said words comprises heuristic analysis of said words in said non-text part using the same heuristic engine.

15. A system as in claim 8, wherein said communication device includes fax hardware.

16. A facsimile apparatus, comprising:

a fax hardware part, having structure to receive facsimile communications; and

a fax contents processor, which analyzes a content of the communications, and determines if the communications is one which likely represents an undesirable communication, wherein said processor operates to obtain a hash of at least a portion of an image representing the facsimile communications, and to compare said hash to plural hashes of known undesirable images in a database to determine undesirable communications based on a match therebetween.

17. An apparatus as in claim 16, wherein said processor operates to prevent the facsimile from being automatically provided based on said determining that the communications is likely undesirable.

18. An apparatus as in claim 17, further comprising a printer that prints facsimile communications, and wherein said prevent comprises printing only communications which are not determined to represent undesirable communications.

19. An apparatus as in claim 16 wherein said fax contents processor processes a file indicative of an image representing the facsimile communication.

20. An apparatus as in claim 19, wherein said image is processed to optically character recognized text within the image, and to process the text to determine words which likely represent undesirable communications.

21. An apparatus as in claim 19, further comprising a memory storing image parts representing parts from known undesirable communications by comparing the image with a catalog of image information that includes known image information therein, where said automatically determining determines multiple said categories, where at least one of the known information represents an undesired category, and wherein said processor processes the image to compare parts of the image to said parts in said memory.