Video comparison using color histograms

• US 9,158,993 B2
• Filed: 11/24/2014
• Issued: 10/13/2015
• Est. Priority Date: 12/13/2011
• Status: Active Grant

• Current Assignee: The Nielsen Company LLC (MOOD Media Corporation)
• Original Assignee: The Nielsen Company LLC (MOOD Media Corporation)
• Inventors: Lee, Morris
• Primary Examiner(s): Dang, Duy M
• Application Number:
  US14/551,828
• Publication Number:
  US 20150078659A1
• Time in Patent Office: 323 Days
• Field of Search: 382/164, 382/165, 382/170, 382/305, 345/88, 345/690, 348/700, 348/722, 725/19, 709/224
• US Class Current: 1/1
• CPC Class Codes: G06F 16/785 using colour or luminescence
• CPC Class Codes: G06V 10/50 by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
• CPC Class Codes: G06V 10/56 relating to colour
• CPC Class Codes: G06V 20/48 Matching video sequences

Abstract

Methods, apparatus and articles of manufacture for video comparison using color histograms are disclosed. Example methods disclosed herein to compare video sequences include determining a color histogram corresponding to an input video sequence based on color values of pixels sampled from a plurality of video frames of the input video sequence. Such example methods also include adjusting the color histogram corresponding to the input video sequence based on a first reference color histogram corresponding to a first reference video sequence to determine a first adjusted color histogram corresponding to the input video sequence. Such example methods further include comparing the adjusted color histogram and the first reference color histogram to determine whether the first reference video sequence matches the input video sequence.

Claims

1. A method to compare video sequences, the method comprising:

determining a color histogram corresponding to an input video sequence based on color values of pixels sampled from a plurality of video frames of the input video sequence;

tiling video frames of a first reference video sequence to determine a composite image;

determining a first reference color histogram corresponding to the first reference video sequence based on color values of pixels in the composite image;

adjusting the color histogram corresponding to the input video sequence based on the first reference color histogram corresponding to the first reference video sequence to determine an adjusted color histogram corresponding to the input video sequence; and

comparing the adjusted color histogram and the first reference color histogram to determine whether the first reference video sequence matches the input video sequence.

2. A method as defined in claim 1, wherein determining the color histogram corresponding to the input video sequence includes:

sampling a set of pixels from the plurality of video frames of the input video sequence;

determining respective color combinations for respective ones of the set of pixels, a first color combination for a first one of the set of pixels being determined by combining a first color value for the first one of the set of pixels and a second color value for a neighboring pixel of the first one of the set of pixels; and

counting respective occurrences of different ones of the color combinations to determine the color histogram corresponding to the input video sequence.

3. A method as defined in claim 2, further including:

capturing the input video sequence using a camera operating at a first frame rate;

sampling the input video sequence at a second frame rate lower than the first frame rate to obtain the plurality of video frames of the input video sequence; and

displaying a result indicating whether the first reference video sequence matches the input video sequence.

4. A method as defined in claim 1, further including:

determining a first comparison value by comparing the adjusted color histogram and the first reference color histogram;

adjusting the first reference color histogram based on the color histogram corresponding to the input video sequence to determine a first adjusted reference color histogram;

determining a second comparison value by comparing the first adjusted reference color histogram and the color histogram corresponding to the input video sequence; and

determining that the first reference video sequence matches the input video sequence when at least one of the first comparison value or the second comparison value satisfies a threshold.

5. A method as defined in claim 1, wherein adjusting the color histogram corresponding to the input video sequence based on the first reference color histogram includes:

multiplying a value of a first bin of the color histogram corresponding to the input video sequence by a scale factor to determine a scaled bin value; and

determining an adjusted value of a corresponding first bin of the adjusted color histogram to be a smaller of the scaled bin value and a value of a corresponding first bin of the reference color histogram.

6. A method as defined in claim 5, wherein the scale factor is adjustable.

7. A tangible computer readable medium comprising computer readable instructions which, when executed, cause a processor to at least:

determine a color histogram corresponding to an input video sequence based on color values of pixels sampled from a plurality of video frames of the input video sequence;

tile video frames of a first reference video sequence to determine a composite image;

determine a first reference color histogram corresponding to the first reference video sequence based on color values of pixels in the composite image;

adjust the color histogram corresponding to the input video sequence based on the first reference color histogram corresponding to the first reference video sequence to determine an adjusted color histogram corresponding to the input video sequence; and

compare the adjusted color histogram and the first reference color histogram to determine whether the first reference video sequence matches the input video sequence.

8. A tangible computer readable medium as defined in claim 7, wherein to determine the color histogram corresponding to the input video sequence, the instructions, when executed, cause the processor to:

sample a set of pixels from the plurality of video frames of the input video sequence;

determine respective color combinations for respective ones of the set of pixels, a first color combination for a first one of the set of pixels being determined by combining a first color value for the first one of the set of pixels and a second color value for a neighboring pixel of the first one of the set of pixels; and

count respective occurrences of different ones of the color combinations to determine the color histogram corresponding to the input video sequence.

9. A tangible computer readable medium as defined in claim 8, wherein the instructions, when executed, further cause the processor to:

capture the input video sequence using a camera operating at a first frame rate;

sample the input video sequence at a second frame rate lower than the first frame rate to obtain the plurality of video frames of the input video sequence; and

display a result indicating whether the first reference video sequence matches the input video sequence.

10. A tangible computer readable medium as defined in claim 7, wherein the instructions, when executed, further cause the processor to:

determine a first comparison value by comparing the adjusted color histogram and the first reference color histogram;

adjust the first reference color histogram based on the color histogram corresponding to the input video sequence to determine a first adjusted reference color histogram;

determine a second comparison value by comparing the first adjusted reference color histogram and the color histogram corresponding to the input video sequence; and

determine that the first reference video sequence matches the input video sequence when at least one of the first comparison value or the second comparison value satisfies a threshold.

11. A tangible computer readable medium as defined in claim 7, wherein to adjust the color histogram corresponding to the input video sequence based on the first reference color histogram, the instructions, when executed, further cause the processor to:

multiply a value of a first bin of the color histogram corresponding to the input video sequence by a scale factor to determine a scaled bin value; and

determine an adjusted value of a corresponding first bin of the adjusted color histogram to be a smaller of the scaled bin value and a value of a corresponding first bin of the reference color histogram.

12. A tangible computer readable medium as defined in claim 11, wherein the scale factor is adjustable.

13. An apparatus to compare video sequences, the apparatus comprising:

a histogram generator to determine a color histogram corresponding to an input video sequence based on color values of pixels sampled from a plurality of video frames of the input video sequence; and

a histogram comparator to;

adjust the color histogram corresponding to the input video sequence based on a first reference color histogram corresponding to a first reference video sequence to determine an adjusted color histogram corresponding to the input video sequence; and

compare the adjusted color histogram and the first reference color histogram to determine whether the first reference video sequence matches the input video sequence;

wherein the histogram comparator includes a bin adjuster to;

multiply a value of a first bin of the color histogram corresponding to the input video sequence by a scale factor to determine a scaled bin value; and

determine an adjusted value of a corresponding first bin of the adjusted color histogram to be a smaller of the scaled bin value and a value of a corresponding first bin of the reference color histogram.

14. An apparatus as defined in claim 13, wherein the a histogram generator includes:

an image sampler to sample a set of pixels from the plurality of video frames of the input video sequence;

a color quantizer to determine respective color combinations for respective ones of the set of pixels, a first color combination for a first one of the set of pixels being determined by combining a first color value for the first one of the set of pixels and a second color value for a neighboring pixel of the first one of the set of pixels; and

a color counter to count respective occurrences of different ones of the color combinations to determine the color histogram corresponding to the input video sequence.

15. An apparatus as defined in claim 14, further including:

a video interface to capture the input video sequence using a camera operating at a first frame rate;

a video frame sampler to sample the input video sequence at a second frame rate lower than the first frame rate to obtain the plurality of video frames of the input video sequence; and

a results interface to display a result indicating whether the first reference video sequence matches the input video sequence.

16. An apparatus as defined in claim 13, wherein the histogram comparator includes:

a bin adjuster to;

determine the adjusted color histogram; and

adjust the first reference color histogram based on the color histogram corresponding to the input video sequence to determine a first adjusted reference color histogram;

a comparison metric determiner to;

determine a first comparison value by comparing the adjusted color histogram and the first reference color histogram; and

determine a second comparison value by comparing the first adjusted reference color histogram and the color histogram corresponding to the input video sequence; and

a threshold comparator to determine that the first reference video sequence matches the input video sequence when at least one of the first comparison value or the second comparison value satisfies a threshold.

17. An apparatus as defined in claim 13, wherein the scale factor is adjustable.