Methods and apparatus for audience measurement using global signature representation and matching

• US 8,707,341 B2
• Filed: 06/25/2012
• Issued: 04/22/2014
• Est. Priority Date: 08/15/2007
• Status: Active Grant

• Current Assignee: Nielsen Media Research, Inc. (Nielsen Holdings plc)
• Original Assignee: The Nielsen Company LLC (MOOD Media Corporation)
• Inventors: Deng, Kevin Keqiang
• Primary Examiner(s): Duong, Frank
• Application Number:
  US13/532,264
• Publication Number:
  US 20120266187A1
• Time in Patent Office: 666 Days
• Field of Search: 375/240.01, 375/240.26, 386/1, 386/45, 386/46, 386/125, 386/126, 725/9, 725/14, 725/16, 725 19- 21, 725/80, 725/85, 725/97, 725/133, 725/141, 725/153
• US Class Current: 725/19
• CPC Class Codes: H04H 60/31 Arrangements for monitoring the use made of the broadcast services
• CPC Class Codes: H04H 60/37 for identifying segments of broadcast information, e.g. scenes or extracting programme ID
• CPC Class Codes: H04H 60/40 for identifying broadcast time
• CPC Class Codes: H04H 60/56 Arrangements characterised by components specially adapted for monitoring, identification or recognition covered by groups H04H60/29-H04H60/54
• CPC Class Codes: H04N 21/44224 Monitoring of user activity on external systems, e.g. Internet browsing

Abstract

Disclosed example methods to determine a time offset between a first media content signature comprising a first sequence of first data pairs and a second media content signature comprising a second sequence of second data pairs include determining a histogram comprising a plurality of bins representative of a plurality of determined time differences. The plurality of determined time differences include each respective time difference determined between each first data pair and second data pair combination. Each respective bin of the histogram represents a respective window of time differences and has a respective value representative of a number of the plurality of determined time differences lying within the respective window of time differences represented by the respective bin. Such disclosed example methods also include selecting at least one bin of the histogram to correspond to the time offset between the first signature and the second signature.

Claims

1. A method to determine a time offset between a first media signature comprising a first sequence of first data pairs and a second media signature comprising a second sequence of second data pairs, the method comprising:

comparing, using a processor, each first data pair in the first sequence with each second data pair in the second sequence to determine a respective time difference between each first data pair and second data pair combination;

determining, using the processor, a histogram comprising a plurality of bins representative of a plurality of determined time differences, the plurality of determined time differences comprising each respective time difference determined between each first data pair and second data pair combination, each respective bin of the histogram representing a respective window of time differences and having a respective value representative of a number of the plurality of determined time differences lying within the respective window of time differences represented by the respective bin; and

selecting at least one bin of the histogram to correspond to the time offset between the first signature and the second signature.

2. A method as defined in claim 1 further comprising:

converting the first signature to have a signature representation wherein each respective first data pair of the first sequence includes a respective first time value representative of a respective elapsed time between the respective first data pair and a starting first data pair of the first sequence; and

converting the second signature to have a signature representation wherein each respective second data pair of the second sequence includes a respective second time value representative of a respective elapsed time between the respective second data pair and a starting second data pair of the second sequence.

3. A method as defined in claim 1 wherein selecting the at least one bin comprises selecting a bin having a maximum value among the plurality of bins to correspond to the time offset between the first signature and the second signature.

4. A method as defined in claim 1 wherein selecting the at least one bin comprises selecting a set of bins to correspond to the time offset between the first signature and the second signature, a respective value of each one of the set of bins meeting a threshold.

5. A method as defined in claim 1 further comprising dividing each of the plurality of determined time differences by a window width to determine a plurality of scaled time differences, and the respective window of time differences represented by each bin corresponds to a respective scaled time difference in the plurality of scaled time differences.

6. A method as defined in claim 1 further comprising determining the value of each respective bin to represent a distance between the first signature and the second signature when the signatures are time-shifted with respect to each other by a shift amount associated with the respective bin.

7. A method as defined in claim 6 further comprising determining the respective value of each respective bin based on a ratio of a number of the plurality of determined time differences lying within the respective window of time differences represented by the respective bin and a total number of data pairs in the first sequence or the second sequence.

8. A method as defined in claim 1 further comprising scaling the selected at least one bin by a ratio based on a second number of the plurality of determined differences lying within the respective window of time differences represented by the bin, wherein each of the plurality of time differences counted in the second number is associated with respective first and second data pairs having a similar magnitude trend.

9. A tangible machine readable storage device or storage disk comprising machine readable instructions which, when executed, cause a machine to at least:

compare each first data pair in a first media signature comprising a first sequence of first data pairs with each second data pair in a second media signature comprising a second sequence of second data pairs to determine a respective time difference between each first data pair and second data pair combination;

determine a histogram comprising a plurality of bins representative of a plurality of determined time differences, the plurality of determined time differences comprising each respective time difference determined between each first data pair and second data pair combination, each respective bin of the histogram representing a respective window of time differences and having a respective value representative of a number of the plurality of determined time differences lying within the respective window of time differences represented by the respective bin; and

select at least one bin of the histogram to correspond to a time offset between the first signature and the second signature.

10. A storage device or storage disk as defined in claim 9 wherein the machine readable instructions, when executed, further cause the machine to:

convert the first signature to have a signature representation wherein each respective first data pair of the first sequence includes a respective first time value representative of a respective elapsed time between the respective first data pair and a starting first data pair of the first sequence; and

convert the second signature to have a signature representation wherein each respective second data pair of the second sequence includes a respective second time value representative of a respective elapsed time between the respective second data pair and a starting second data pair of the second sequence.

11. A storage device or storage disk as defined in claim 9 wherein to select the at least one bin, the machine readable instructions, when executed, further cause the machine to select a bin having a maximum value among the plurality of bins to correspond to the time offset between the first signature and the second signature.

12. A storage medium as defined in claim 9 wherein to select the at least one bin, the machine readable instructions, when executed, further cause the machine to select a set of bins to correspond to the time offset between the first signature and the second signature, a respective value of each one of the set of bins meeting a threshold.

13. A storage device or storage disk as defined in claim 9 wherein the machine readable instructions, when executed, further cause the machine to divide each of the plurality of determined time differences by a window width to determine a plurality of scaled time differences, and the respective window of time differences represented by each bin corresponds to a respective scaled time difference in the plurality of scaled time differences.

14. A storage device or storage disk as defined in claim 9 wherein the machine readable instructions, when executed, further cause the machine to scale the selected at least one bin by a ratio based on a second number of the plurality of determined differences lying within the respective window of time differences represented by the bin, wherein each of the plurality of time differences counted in the second number is associated with respective first and second data pairs having a similar magnitude trend.

15. An apparatus comprising:

a processor to;

compare each first data pair in a first media signature comprising a first sequence of first data pairs with each second data pair in a second media signature comprising a second sequence of second data pairs to determine a respective time difference between each first data pair and second data pair combination;

determine a histogram comprising a plurality of bins representative of a plurality of determined time differences, the plurality of determined time differences comprising each respective time difference determined between each first data pair and second data pair combination, each respective bin of the histogram representing a respective window of time differences and having a respective value representative of a number of the plurality of determined time differences lying within the respective window of time differences represented by the respective bin; and

select at least one bin of the histogram to correspond to a time offset between the first signature and the second signature; and

a memory to store the histogram.

16. An apparatus as defined in claim 15 wherein the processor is further to:

convert the first signature to have a signature representation wherein each respective first data pair of the first sequence includes a respective first time value representative of a respective elapsed time between the respective first data pair and a starting first data pair of the first sequence; and

convert the second signature to have a signature representation wherein each respective second data pair of the second sequence includes a respective second time value representative of a respective elapsed time between the respective second data pair and a starting second data pair of the second sequence.

17. An apparatus as defined in claim 15 wherein to select the at least one bin, the processor is further to select a bin having a maximum value among the plurality of bins to correspond to the time offset between the first signature and the second signature.

18. An apparatus as defined in claim 15 wherein to select the at least one bin, the processor is further to select a set of bins to correspond to the time offset between the first signature and the second signature, a respective value of each one of the set of bins meeting a threshold.

19. An apparatus as defined in claim 15 wherein the processor further is to divide each of the plurality of determined time differences by a window width to determine a plurality of scaled time differences, and the respective window of time differences represented by each bin corresponds to a respective scaled time difference in the plurality of scaled time differences.

20. An apparatus as defined in claim 15 wherein the processor further is to scale the selected at least one bin by a ratio based on a second number of the plurality of determined differences lying within the respective window of time differences represented by the bin, wherein each of the plurality of time differences counted in the second number is associated with respective first and second data pairs having a similar magnitude trend.