System for identifying content of digital data

• US 8,112,818 B2
• Filed: 10/24/2007
• Issued: 02/07/2012
• Est. Priority Date: 07/27/2007
• Status: Active Grant

• Current Assignee: Audible Magic Corporation
• Original Assignee: Audible Magic Corporation
• Inventors: Wold, Erling
• Primary Examiner(s): Chai, Longbit
• Application Number:
  US11/923,491
• Publication Number:
  US 20090031326A1
• Time in Patent Office: 1,567 Days
• Field of Search: 726/32
• US Class Current: 726/32
• CPC Class Codes: G06F 16/7834 using audio features
• CPC Class Codes: G06F 16/785 using colour or luminescence
• CPC Class Codes: G06F 21/10 Protecting distributed programs or content, e.g. vending or licensing of copyrighted material protection in video systems or pay television H04N7/16 ; Digital rights management [DRM]
• CPC Class Codes: G06F 21/16 Program or content traceability, e.g. by watermarking

Abstract

A system for identifying a work represented by digital data. This system detects events in data of the content of an unknown work and measures a metric between the events. The system then compares the detected events and metrics from an unknown work with the events and metrics of a known work. If the events and metrics match, the unknown work is identified as a copy of the known work.

Claims

1. A method, implemented by a computing system programmed to perform the following, comprising:

selecting, by the computing system, a portion of data of an unknown work;

detecting, by the computing system, each event in said portion of data of said unknown work, wherein an event is a perceptual occurrence in a work that can be successively positioned in time;

determining, by the computing system, an event metric between each successive event in said portion of data in said unknown work;

generating, by the computing system, a first list of event metrics between said events for said unknown work;

receiving, by the computing system, a second list of event metrics for a known work;

generating, by the computing system, a third list comprising events in the first list that match events in the second list by

performing a comparison to determine whether an Mth event of the first list matches an Nth event of the second list, wherein M is an index over events in the first list and N is an index over events in the second list;

responsive to a determination of a match, adding the matching events to the third list and incrementing the index N;

responsive to a determination that there is not a match, incrementing the index M,

wherein the comparison is performed until the index N is greater than or equal to a number of events in the second list and the index M is greater than a threshold; and

determining said unknown work is a copy of said known work based on the third list comprising the events in the first list that match events in the second list.

2. The method of claim 1, wherein the threshold is a minimum of 3 and the number of events in the second list.

3. The method of claim 1 wherein determining said unknown work is a copy of said known work comprises:

generating the third list wherein each M+X event of said first list for said unknown work is paired with the N+X event of said second list for said known work; and

computing a regression line through said associated pairs in said list of associated pairs.

4. The method of claim 3 further comprising:

computing an error tolerance from said regression line.

5. The method of claim 3 further comprising:

determining a regression error from said computation of said regression line;

comparing said regression error to a regression error threshold; and

determining said unknown work is a copy of said known work responsive to said regression error being less than said regression error threshold.

6. The apparatus of claim 3 further comprising:

determining a weighted error from said computation of said regression line;

comparing said weighted error to a weighted error threshold; and

determining said unknown work is a copy of said known work responsive to said weighted error being less than said weighted error threshold.

7. The method of claim 3 further comprising:

determining a miss ratio from said computation of said regression line;

comparing said miss ratio to a miss ratio threshold; and

determining said unknown work is a copy of said known work responsive to said miss ratio being less than said miss ratio threshold.

8. The method of claim 3 further comprising:

determining a weighted miss ratio from said computation of said regression line;

comparing said weighted miss ratio to a weighted miss ratio threshold; and

determining said unknown work is a copy of said known work responsive to said weighted miss ratio being less than said weighted miss ratio threshold.

9. A system comprising:

a memory; and

a processor coupled with the memory to

select a portion of data of an unknown work;

detect each event in said portion of data of said unknown work, wherein an event is a perceptual occurrence in a work that can be successively positioned in time;

determine an event metric between each successive event in said portion of data in said unknown work;

generate a first list of event metrics between said events for said unknown work;

receive a second list of event metrics for a known work;

generate a third list comprising events in the first list that match events in the second list; and

determine said unknown work is a copy of said known work based on the third list comprising the events in the first list that match events in the second list,

wherein to generate the third list comprises the processor to

perform a comparison to determine whether an Mth event of the first list matches an Nth event of the second list, wherein M is an index over events in the first list and N is an index over events in the second list, add the matching events to the third list and incrementing the index N responsive to a determination of a match, and responsive to a determination that there is not a match, increment the index M, wherein the comparison is performed until the index N is no longer less than a number of events in the second list and the index M is greater than a threshold.

10. The system of claim 9, wherein the threshold is a minimum of 3 and the number of events in the second list.

11. The system of claim 9, wherein determining said unknown work is a copy of said known work comprises the processor to:

generate the third list wherein each M+X event of said first list for said unknown work is paired with the N+X event of said second list for said known work; and

compute a regression line through said associated pairs in said list of associated pairs.

12. The system of claim 11, further comprising the processor to:

compute an error tolerance from said regression line.

13. The system of claim 11, further comprising the processor to:

determine a regression error from said computation of said regression line;

compare said regression error to a regression error threshold; and

determine said unknown work is a copy of said known work responsive to said regression error being less than said regression error threshold.

14. The system of claim 11, further comprising the processor to:

determine a weighted error from said computation of said regression line;

compare said weighted error to a weighted error threshold; and

determine said unknown work is a copy of said known work responsive to said weighted error being less than said weighted error threshold.

15. The system of claim 11, further comprising the processor to:

determine a miss ratio from said computation of said regression line;

compare said miss ratio to a miss ratio threshold; and

determine said unknown work is a copy of said known work responsive to said miss ratio being less than said miss ratio threshold.

16. The system of claim 11, further comprising the processor to:

determine a weighted miss ratio from said computation of said regression line;

compare said weighted miss ratio to a weighted miss ratio threshold; and

determine said unknown work is a copy of said known work responsive to said weighted miss ratio being less than said weighted miss ratio threshold.

17. A non-transitory computer readable storage medium that provides instructions, which when executed on a processing system cause the processing system to perform a method comprising:

selecting, by the processing system, a portion of data of an unknown work;

detecting, by the processing system, each event in said portion of data of said unknown work, wherein an event is a perceptual occurrence in a work that can be successively positioned in time;

determining, by the processing system, an event metric between each successive event in said portion of data in said unknown work;

generating, by the processing system, a first list of event metrics between said events for said unknown work;

receiving, by the processing system, a second list of event metrics for a known work;

generating, by the processing system, a third list comprising events in the first list that match events in the second list by

performing a comparison to determine whether an Mth event of the first list matches an Nth event of the second list;

responsive to a determination of a match, adding the matching events to the third list and incrementing the index N;

responsive to a determination that there is not a match, incrementing the index M;

wherein the comparison is performed until the index N is no longer less than a number of events in the second list and the index M is greater than a threshold; and

determining, by the processing system, said unknown work is a copy of said known work based on the third list comprising the events in the first list that match events in the second list.

18. The non-transitory computer readable storage medium of claim 17, wherein the threshold is a minimum of 3 and the number of events in the second list.

19. The non-transitory computer readable storage medium of claim 17, wherein determining said unknown work is a copy of said known work comprises:

generating the third list wherein each M+X event of said first list for said unknown work is paired with the N+X event of said second list for said known work; and

computing a regression line through said associated pairs in said list of associated pairs.

20. The non-transitory computer readable storage medium of claim 19, further comprising:

computing an error tolerance from said regression line.

21. The non-transitory computer readable storage medium of claim 19, further comprising:

determining a regression error from said computation of said regression line;

comparing said regression error to a regression error threshold; and

determining said unknown work is a copy of said known work responsive to said regression error being less than said regression error threshold.

22. The non-transitory computer readable storage medium of claim 19, further comprising:

determining a weighted error from said computation of said regression line;

comparing said weighted error to a weighted error threshold; and

determining said unknown work is a copy of said known work responsive to said weighted error being less than said weighted error threshold.

23. The non-transitory computer readable storage medium of claim 19, further comprising:

determining a miss ratio from said computation of said regression line;

comparing said miss ratio to a miss ratio threshold; and

determining said unknown work is a copy of said known work responsive to said miss ratio being less than said miss ratio threshold.

24. The non-transitory computer readable storage medium of claim 19, further comprising:

determining a weighted miss ratio from said computation of said regression line;

comparing said weighted miss ratio to a weighted miss ratio threshold; and

determining said unknown work is a copy of said known work responsive to said weighted miss ratio being less than said weighted miss ratio threshold.