Methods and apparatus to count people in images

• US 9,275,285 B2
• Filed: 03/29/2012
• Issued: 03/01/2016
• Est. Priority Date: 03/29/2012
• Status: Active Grant

• Current Assignee: The Nielsen Company LLC (MOOD Media Corporation)
• Original Assignee: The Nielsen Company LLC (MOOD Media Corporation)
• Inventors: Srinivasan, Venugopal, Soundararajan, Padmanabhan
• Primary Examiner(s): Vu, Kim
• Assistant Examiner(s): VANCHY JR, MICHAEL J
• Application Number:
  US13/434,330
• Publication Number:
  US 20130259380A1
• Time in Patent Office: 1,433 Days
• Field of Search: 382/103, 382/118
• US Class Current: 1/1
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: G06V 40/16 Human faces, e.g. facial parts, sketches or expressions

Abstract

Methods and apparatus to count people in images are disclosed. An example method includes generating a first fluctuation factor for a first frame by averaging fluctuation values of a random set of pixels of the first frame; generating a second fluctuation factor for a person indication area of the first frame by averaging fluctuation values of pixels of the person indication area; and marking the person indication area as a false positive when the second fluctuation factor is less than or equal to the first fluctuation factor.

Claims

1. A method, comprising:

generating, with a random number generator, a random subset of pixels of a first frame, the random subset of pixels including less than all of the pixels of the first frame;

generating, via a processor, a first fluctuation factor for the first frame by averaging fluctuation values of the random subset of pixels of the first frame;

identifying, via the processor, a person indication area in which a person is believed to be located, the person indication area corresponding to a subset of the pixels of the first frame including less than all of the pixels of the first frame;

generating, via the processor, a second fluctuation factor for the person indication area of the first frame by averaging fluctuation values of the subset of the pixels corresponding to the person indication area; and

marking, via the processor, the person indication area as corresponding to a false positive when the second fluctuation factor is less than or equal to the first fluctuation factor.

2. A method as defined in claim 1, wherein identifying the person indication area further includes calculating a location of the person indication area in the first frame.

3. A method as defined in claim 2, further including querying a database with the calculated location to determine whether a person was detected at the calculated location in previous frames captured within a threshold amount of time from capture of the first frame.

4. A method as defined in claim 3, further including unmarking the person indication area as a false positive when the query indicates that at least one of the previous frames includes a detection of a person at the calculated location.

5. A method as defined in claim 3, further including verifying the person indication area as a false positive when the query indicates that the previous frames do not include a detection of a person at the calculated location.

6. A method as defined in claim 1, wherein the generating of the first fluctuation factor includes performing root mean square calculations.

7. A method as defined in claim 6, wherein the root mean square calculations are based on pixel intensities.

8. A tangible computer readable storage medium comprising instructions that, when executed, cause a machine to at least:

generate a random subset of pixels of a first frame, the random subset of pixels including less than all of the pixels of the first frame;

generate a first fluctuation factor for the first frame by averaging fluctuation values of the random subset of pixels of the first frame;

identify a person indication area in which a person is believed to be located, the person indication area corresponding to a subset of the pixels of the first frame including less than all of the pixels of the first frame;

generate a second fluctuation factor for the person indication area of the first frame by averaging fluctuation values of the subset of the pixels corresponding to the person indication area; and

identify the person indication area as corresponding to a false positive when the second fluctuation factor is less than or equal to the first fluctuation factor.

9. A tangible computer readable storage medium as defined in claim 8, wherein the instructions, when executed, cause the machine to calculate a location of the person indication area in the first frame.

10. A tangible computer readable storage medium as defined in claim 9, wherein the instructions, when executed, cause the machine to query a database with the calculated location to determine whether a person was detected at the calculated location in previous frames captured within a threshold amount of time from capture of the first frame.

11. A tangible computer readable storage medium as defined in claim 10, wherein the instructions, when executed, cause the machine to reverse the identification of the person indication area as a false positive when the query indicates that at least one of the previous frames includes a detection of a person at the calculated location.

12. A tangible computer readable storage medium as defined in claim 10, wherein the instructions, when executed, cause the machine to verify the person indication area as a false positive when the query indicates that the previous frames do not include a detection of a person at the calculated location.

13. A tangible computer readable storage medium as defined in claim 8, wherein the instructions cause the machine to generate the first fluctuation factor by performing root mean square calculations.

14. A tangible computer readable storage medium as defined in claim 13, wherein the root mean square calculations are based on pixel intensities.

15. An apparatus, comprising:

a processor to;

generate a random subset of pixels of a first frame, the random subset of pixels including less than all of the pixels of the first frame;

generate a first fluctuation factor for the first frame by averaging fluctuation values of the random subset of pixels of the first frame; and

identify a person indication area in which a person is believed to be located, the person indication area corresponding to a subset of the pixels of the first frame including less than all of the pixels of the first frame;

generate a second fluctuation factor for the person indication area of the first frame by averaging fluctuation values of the subset of the pixels corresponding to the person indication area; and

a comparator to identify the person indication area as corresponding to a false positive when the second fluctuation factor is less than or equal to the first fluctuation factor.

16. An apparatus as defined in claim 15, further including a location calculator to determine a location of the person indication area in the first frame.

17. An apparatus as defined in claim 16, further including a checker to query a database with the calculated location to determine whether a person was detected at the calculated location in previous frames captured within a threshold amount of time from capture of the first frame.

18. An apparatus as defined in claim 17, wherein the checker is to reverse the identification of the person indication area as a false positive when the query indicates that at least one of the previous frames includes a detection of a person at the calculated location.

19. An apparatus as defined in claim 17, wherein the checker is to verify the person indication area as a false positive when the query indicates that the previous frames do not include a detection of a person at the calculated location.

20. An apparatus as defined in claim 15, wherein the processor is to generate the first fluctuation via root mean square calculations based on pixel intensities.