Methods and apparatus for identifying media content using temporal signal characteristics
Methods and apparatus to identify media content using temporal signal characteristics are disclosed. An example method includes generating a reference signature based on a reference signal corresponding to known media, generating sums based on peaks in a media signal corresponding to media, identifying signal peaks based on the generated sums, generating a second signature based on normalized curve features, wherein the normalized curve features respectively correspond to the identified signal peaks at a corresponding temporal locations of the corresponding signal peak, and determining whether the media signal corresponds to the reference signal based on a comparison of the reference signature and the second signature.
1. An apparatus comprising:
a peak detector to detect peaks of a sensed media signal during time intervals and to output absolute values of the detected peaks;
a summer to generate interval sums respective to the time intervals based on the absolute values;
a peak identifier to identify second peaks based on the interval sums; and
a feature generator to generate and output a signature representative of the media signal, the signature including normalized curve features located at temporal locations of the second peaks, at least one of the peak detector, the summer, the peak identifier or the feature generator implemented with hardware or at least one processor.
2. The apparatus of claim 1 , wherein the signature is a first signature, further including a correlator to:
compare the first signature representative of the media signal to a second signature representative of a reference media signal; and
generate an index based on the comparison of the first signature and the second signature.
3. The apparatus of claim 2 , further including an index comparator to:
compare the index to a threshold; and
determine whether the media signal matches the reference media signal based on the comparison of the index to the threshold.
4. An apparatus comprising:
a peak detector to detect peaks of a sensed media signal during time intervals and to output absolute values of the detected peaks;
a summer to generate interval sums respective to the time intervals based on the absolute values;
a peak identifier to identify second peaks based on the interval sums; and
a feature generator to generate and output a signature representative of the media signal, the signature including normalized curve features based on the second peaks, the normalized curve features having equal heights, at least one of the peak detector, the summer, the peak identifier or the feature generator implemented with hardware or at least one processor.
5. The apparatus of claim 4 , wherein the normalized curve features have equal widths.
6. An apparatus comprising:
a peak detector to detect peaks of a sensed media signal during time intervals and to output absolute values of the detected peaks;
a summer to generate interval sums respective to the time intervals based on the absolute values;
a peak identifier to identify second peaks based on the interval sums; and
a feature generator to generate and output a signature representative of the media signal, the signature including normalized curve features based on the second peaks, the signature having a substantially zero value except at temporal locations corresponding to the normalized curve features, at least one of the peak detector, the summer, the peak identifier or the feature generator implemented with hardware or at least one processor.
7. The apparatus of claim 1 , further including a smoother to smooth the interval sums.
8. A method comprising:
detecting, by executing one or more instructions with a processor, peaks in time intervals of a sensed media signal;
generating, by executing one or more instructions with the processor, interval sums respective to the time intervals based on absolute values of the detected peaks;
generating, by executing one or more instructions with the processor, a curve based on the interval sums;
generating, by executing one or more instructions with the processor, normalized curve features at respective temporal locations of peaks in the curve; and
outputting, by executing one or more instructions with the processor, a signature based on the normalized curve features, the signature having a substantially zero value except at the normalized curve features, the signature representative of the media signal.
9. The method of claim 8 , wherein the signature is a first signature, and further including determining whether the media signal corresponds to a reference media signal based on a comparison of a reference signature and the first signature.
10. The method of claim 9 , further including generating an index value based on the comparison, wherein the determining of whether the sensed media signal corresponds to the reference media signal includes determining whether the index value satisfies a threshold.
11. The method of claim 8 , wherein the normalized curve features have respective heights, widths, and center locations.
12. The method of claim 8 , further including smoothing the curve prior to generating the normalized curved features.
13. A method comprising:
detecting, by executing one or more instructions with a processor, peaks in time intervals of a sensed media signal;
generating, by executing one or more instructions with the processor, interval sums respective to the time intervals based on absolute values of the detected peaks;
generating, by executing one or more instructions with the processor, a curve based on the interval sums;
generating, by executing one or more instructions with the processor, normalized curve features at respective temporal locations of peaks in the curve, the normalized curved features have heights, widths, and center locations, the heights of the normalized curve features being equal and the widths of the normalized curve features being equal; and
outputting, by executing one or more instructions with the processor, a signature based on the normalized curve features, the signature representative of the media signal.
14. A computer readable storage device or storage disk comprising instructions that, when executed, cause a processor to at least:
identify peaks in time intervals of a sensed media signal;
determine interval sums respective to the time intervals based on absolute values of the peaks occurring in the time intervals;
identify second peaks based on the interval sums; and
generate a signature to identify the media signal, the signature including normalized curve features located at temporal locations of the second peaks.
15. The computer readable storage device or storage disk of claim 14 , wherein the instructions, when executed, further cause the processor to generate an index based on a comparison of the signature generated to identify the media signal and a reference signature associated with a reference media signal.
16. The computer readable storage device or storage disk of claim 15 , wherein the instructions, when executed, further cause the processor to identify the media signal based on whether the index satisfies a threshold.
17. The computer readable storage device or storage disk of claim 14 , wherein the instructions, when executed, further cause the processor to smooth the interval sums prior to identification of the second peaks.
18. The computer readable storage device or storage disk of claim 14 , wherein the normalized curved features have equal heights.
19. The computer readable storage device or storage disk of claim 14 , wherein the normalized curved features have equal widths.
20. The computer readable storage device or storage disk of claim 14 , wherein the signature has a substantially zero value except at the normalized curve features.