Methods and apparatus to detect commercial advertisements associated with media presentations

Patent metadata

Field	Value
Publication number	US-10028014-B2
Application number	US-201715449160-A
Country	US
Kind code	B2
Filing date	Mar 3, 2017
Priority date	Dec 31, 2009
Publication date	Jul 17, 2018
Grant date	Jul 17, 2018

Abstract

Methods and apparatus to detect commercial advertisements associated with media presentations are disclosed. An example method includes comparing an amount of audio randomness in an inter-segment audio frame to a random level threshold, the amount of audio randomness being based on a random-factor ratio; and in response to the amount of randomness satisfying the random level threshold, indicating the inter-segment audio frame is associated with the commercial transition.

First claim

What is claimed is:

1. A method to detect a commercial transition associated with a media presentation, the method comprising:
comparing, by executing an instruction with a processor, an amount of audio randomness in an inter-segment audio frame to a random level threshold, the amount of audio randomness being based on a random-factor ratio; and
in response to the amount of randomness satisfying the random level threshold, indicating, by executing an instruction with the processor, the inter-segment audio frame is associated with the commercial transition.

2. The method as defined in claim 1 , further including determining the random-factor ratio by:
determining a quantity of unique time-based distances between at least one of a plurality of amplitude peaks or a plurality of amplitude troughs in the inter-segment audio frame; and
determining the random-factor ratio by dividing the quantity of unique time-based distances by the quantity of total time-based distances between the at least one of the amplitude peaks or the amplitude troughs.

3. The method as defined in claim 1 , wherein the commercial transition includes at least one of a transition between two commercial advertisements or a transition between a commercial advertisement and a media program.

4. The method as defined in claim 1 , wherein the inter-segment audio frame is reconstructed from at least one of a television transmission, an internet-streamed video, or a video on demand video.

5. The method as defined in claim 1 , wherein the amount of audio randomness of the inter-segment audio frame is indicative of an inter-segment blank frame occurring between two media program segments.

6. The method as defined in claim 1 , further including:
determining a root-mean-squared value of audio in the inter-segment audio frame; and
determining the amount of audio randomness of the inter-segment audio frame in response to the root-mean-squared value of the audio exceeding a first threshold and being less than a second threshold.

7. The method as defined in claim 1 , further including:
in response to the amount of audio randomness satisfying the random level threshold, generating a first hint value based on the amount of audio randomness, the amount of audio randomness being a first characteristic associated with the media presentation indicative of a possible commercial transition;
generating a second hint value in response to detecting a second characteristic associated with the media presentation indicative of the possible commercial transition;
weighting the first and second hint values;
determining an aggregate weighted hint value based on the first and second weighted hint values; and
confirming a presence of the commercial transition associated with the media presentation based on the aggregate weighted hint value.

8. The method as defined in claim 7 , wherein the second characteristic is a change in box-formatting between consecutive video frames.

9. The method as defined in claim 8 , further including adding a third weighted hint value to the aggregate weighted hint value, the third weighted hint value associated with detecting, in association with the media presentation, at least one of a frame aspect ratio change, an audio format change, a timestamp discontinuity, a closed-captioning text discontinuity, or a commercial insertion opportunity code.

10. A method as defined in claim 7 , wherein the first hint value is weighted based on a reliability of the first characteristic as an indicator of an actual commercial transition and the second hint value is weighted based on the reliability of the second characteristic as an indicator of an actual commercial transition.

11. An apparatus to detect a commercial transition associated with a media presentation, the apparatus comprising:
an audio features detector to compare an amount of audio randomness in an inter-segment audio frame to a random level threshold, the amount of audio randomness being based on a random-factor ratio; and
an advertisement detector to indicate, in response to the amount of audio randomness satisfying the random level threshold, the inter-segment audio frame as occurring at the commercial transition.

12. The apparatus as defined in claim 11 , wherein the audio features detector determines the random-factor ratio by:
determining a quantity of unique time-based distances between at least one of a plurality of amplitude peaks or a plurality of amplitude troughs in the inter-segment audio frame; and
determining the random-factor ratio by dividing the quantity of unique time-based distances by a quantity of total time-based distances between the at least one of the amplitude peaks or the amplitude troughs.

13. The apparatus as defined in claim 11 , wherein the commercial transition includes at least one of a transition between two commercial advertisements or a transition between a commercial advertisement and a media program.

14. The apparatus as defined in claim 11 , wherein the audio frame is reconstructed from at least one of a television transmission, an internet-streamed video, or a video on demand video.

15. A non-transitory computer readable medium comprising instructions that, when executed, cause a processor to at least:
compare an amount of audio randomness in an inter-segment audio frame of a media presentation to a random level threshold, the amount of audio randomness being based on a random-factor ratio; and
in response to the amount of audio randomness satisfying the random level threshold, indicate the inter-segment audio frame is associated with the commercial transition.

16. The non-transitory computer readable medium as defined in claim 15 , wherein the instructions further cause the processor to determine the amount of audio randomness by:
determining a quantity of unique time-based distances between at least one of a plurality of amplitude peaks or a plurality of amplitude troughs in the inter-segment audio frame; and
determining a random-factor ratio by dividing the quantity of unique time-based distances by a quantity of total time-based distances between the at least one of the amplitude peaks or the amplitude troughs.

17. The non-transitory computer readable medium as defined in claim 15 , wherein the audio frame is reconstructed from at least one of a television transmission, an internet-streamed video, or a video on demand video.

18. The non-transitory computer readable medium as defined in claim 15 , wherein the instructions further cause the processor to determine the amount of audio randomness of the audio frame in response to detecting quiet audio in the audio frame.

19. The non-transitory computer readable medium as defined in claim 15 , wherein the amount of audio randomness of the audio frame is indicative of an inter-scene blank frame occurring in a media program or an inter-segment blank frame occurring between two media program segments.

20. The non-transitory computer readable medium as defined in claim 15 , wherein the instructions further cause the processor to:
in response to the amount of audio randomness satisfying the random level threshold, generate a first hint value based on the amount of audio randomness, the amount of audio randomness being a first characteristic associated with the media presentation indicative of a possible commercial transition;
generate a second hint value in response to detecting a second characteristic associated with the media presentation indicative of the possible commercial transition;
weight the

Assignees

• Nielsen Co Us Llc

Inventors

• Wright David H
• Wilson Scott
• Schwerer Ron
• Heffernan Ronan

Classifications

• H04N5/147

  Scene change detection · CPC title

• H04N21/4394 Primary

  involving operations for analysing the audio stream, e.g. detecting features or characteristics in audio streams (arrangements characterised by components specially adapted for monitoring, identification or recognition of audio in broadcast systems H04H60/58) · CPC title

• H04H60/58

  of audio {(determination or detection of speech characteristics in general G10L25/00; speech recognition in general G10L15/00)} · CPC title

• H04H60/48

  for recognising items expressed in broadcast information · CPC title

• H04N21/812

  involving advertisement data (advertising per se G06Q30/02) · CPC title