Methods and apparatus to count people in images

• US 9,594,961 B2
• Filed: 06/05/2015
• Issued: 03/14/2017
• 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): Mistry, Oneal R
• Application Number:
  US14/732,107
• Publication Number:
  US 20150269443A1
• Time in Patent Office: 648 Days
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30242 Counting objects in image
• CPC Class Codes: G06T 7/20 Analysis of motion motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51
• CPC Class Codes: G06T 7/70 Determining position or orientation of objects or cameras camera calibration G06T7/80
• 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
• CPC Class Codes: G06V 40/161 Detection; Localisation; Normalisation

Abstract

Methods and apparatus to count people in images are disclosed. An example method includes identifying, via a processor, active segments of a plurality of segments of a first frame of image data based on respective fluctuation factors; determining, via the processor, that a first one of the active segments is adjacent to or overlaps with a second one of the active segments; and providing, via the processor, a region of interest in the first frame of image data to an object detection operation, the region of interest including the first and second ones of the active segments and excluding an inactive one of the plurality of segments of the first frame.

Claims

1. A method, comprising:

identifying, by executing an instruction with a processor, active segments and inactive segments of a plurality of segments of a first frame of image data by comparing respective fluctuation factors determined for the respective segments to a fluctuation threshold, the fluctuation threshold being calculated based on an average fluctuation of a set of pixels in the first frame of image data;

determining, by executing an instruction with the processor, that a first one of the active segments of the first frame is adjacent to or overlaps with a second one of the active segments of the first frame; and

providing, by executing an instruction with the processor, a region of interest in the first frame of image data to an object detection operation, the region of interest including the first and second ones of the active segments of the first frame and excluding an inactive segment of the first frame.

2. The method as defined in claim 1, wherein the object detection operation does not process the inactive segments.

3. The method as defined in claim 1, further including linking the first and second ones of the active segments to form a shape.

4. The method as defined in claim 3, wherein the shape is a rectangle.

5. The method as defined in claim 1, wherein the fluctuation factors are root mean square values based on pixel intensities.

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

identify active segments and inactive segments of a plurality of segments of a first frame of image data by comparing fluctuation factors determined for the respective segments to a fluctuation threshold, the fluctuation threshold being calculated based on an average fluctuation of a set of pixels in the first frame of image data;

determine that a first one of the active segments of the first frame is adjacent to or overlaps with a second one of the active segments of the first frame; and

provide a region of interest in the first frame of image data to an object detection operation, the region of interest including the first and second ones of the active segments of the first frame and excluding an inactive segment of the first frame.

7. The tangible computer readable storage medium as defined in claim 6, further including instructions to execute the object detection operation, the object detection operation does not process the inactive segments.

8. The tangible computer readable storage medium as defined in claim 6, wherein the instructions, when executed, cause the machine to link the first and second ones of the active segments to form a shape.

9. The tangible computer readable storage medium as defined in claim 8, wherein the shape is a rectangle.

10. The tangible computer readable storage medium as defined in claim 6, wherein the fluctuation factors are root mean square values based on pixel intensities.

11. An apparatus, comprising:

a comparator to identify active segments and inactive segments of a first frame of image data by comparing fluctuation factors determined for respective segments of the first frame to a fluctuation threshold, the fluctuation threshold being calculated based on an average fluctuation of a set of pixels in the first frame of image data;

a segment linker to;

determine that a first one of the active segments of the first frame is adjacent to or overlaps with a second one of the active segments of the first frame; and

define a region of interest in the first frame of image data to include the first and second ones of the active segments of the first frame and exclude an inactive segment of the first frame; and

a detector to perform an object detection operation on the region of interest, at least one of the comparator, the segment linker or the detector implemented via a processor.

12. The apparatus as defined in claim 11, wherein the detector is to not perform the object detection operation on the inactive segment of the first frame.

13. The apparatus as defined in claim 11, wherein the segment linker is to define the region of interest in the first frame of image data by linking the first and second ones of the active segments to form a shape.

14. The apparatus as defined in claim 13, wherein the shape is a rectangle.

15. The apparatus as defined in claim 11, wherein the fluctuation factors are root mean square values based on pixel intensities.