Methods and apparatus to optimize reference signature matching using watermark matching
Methods, apparatus, systems and articles of manufacture are disclosed to optimize reference signature matching using watermark matching. An example disclosed herein includes selecting first unhashed reference signatures to evaluate for a second time segment of monitoring data based on watermark coverage of a first time segment of the monitoring data, the monitoring data obtained from a meter monitoring media presented by a media device and comparing the first unhashed reference signatures with first unhashed monitored signatures from the second time segment of the monitoring data. The example further includes, when a first media event associated with the monitoring data is not identifiable based on the comparing of the first of unhashed reference signatures with the first unhashed monitored signatures hashing the first unhashed monitored signatures to form a corresponding first hashed monitored signatures and comparing the first hashed monitored signatures with a library of reference hashed signatures to determine the first media event associated with the second time segment of the monitoring data.
1. An apparatus, comprising:
an unhashed signature reference determiner to select, from a database, first unhashed reference signatures to evaluate for a second time segment of monitoring data based on watermark coverage of a first time segment of the monitoring data, the monitoring data obtained from a meter monitoring media presented by a media device;
an unhashed signature matcher to compare the first unhashed reference signatures with first unhashed monitored signatures from the second time segment of the monitoring data;
a signature hasher to hash the first unhashed monitored signatures to form corresponding first hashed monitored signatures when a first media event associated with the monitoring data is not identifiable based on the comparison of the first unhashed reference signatures with the first unhashed monitored signatures;
a hashed signature matcher to compare the first hashed monitored signatures with a library of reference hashed signatures to, when the first media event associated with the monitoring data is not identifiable based on the comparison of the first unhashed reference signatures with the first unhashed monitored signatures, identify the first media event associated with the second time segment of the monitoring data; and
a creditor interface to output media identification data based on the first media event.
2. The apparatus of claim 1 , wherein the signature hasher is not to hash the first unhashed monitored signatures when the first media event associated with the monitoring data is identifiable based on the comparison of the first unhashed reference signatures with the first unhashed monitored signatures.
3. The apparatus of claim 1 , wherein the creditor interface to credit a user associated with the monitoring data based on the first media event.
4. The apparatus of claim 1 , further including a watermark matcher to:
determine the watermark coverage of the first time segment based on a presence of a watermark in the monitoring data; and
compare the watermark with a reference watermark to determine a second media event associated with the first time segment of the monitoring data.
5. The apparatus of claim 4 , wherein the unhashed signature reference determiner is to select the first unhashed reference signatures based on an offset between a first timestamp associated with the watermark and a second timestamp of the first unhashed reference signatures.
6. The apparatus of claim 4 , wherein the unhashed signature reference determiner is to select the first unhashed reference signatures based on a media source associated with the reference watermark.
7. The apparatus of claim 1 , wherein:
the unhashed signature reference determiner is to select second unhashed reference signatures to evaluate for a third time segment of monitoring data based on the comparison of the first unhashed reference signatures with first unhashed monitored signatures from the second time segment of the monitoring data; and
the unhashed signature matcher is to compare the second unhashed reference signatures with second unhashed monitored signatures from the second time segment of the monitoring data to identify a third media event associated with the third time segment.
8. The apparatus of claim 1 , wherein:
the unhashed signature reference determiner is to select third unhashed reference signatures to evaluate for a fourth time segment of monitoring data based on the comparison of the first hashed monitored signatures with a library of reference hashed signatures; and
the unhashed signature matcher is to compare the third unhashed reference signatures with third unhashed monitored signatures from the fourth time segment of the monitoring data to identify a fourth media event associated with the fourth time segment.
9. A method, comprising:
selecting, from a database, first unhashed reference signatures to evaluate for a second time segment of monitoring data based on watermark coverage of a first time segment of the monitoring data, the monitoring data obtained from a meter monitoring media presented by a media device;
comparing the first unhashed reference signatures with first unhashed monitored signatures from the second time segment of the monitoring data;
when a first media event associated with the monitoring data is not identifiable based on the comparing of the first unhashed reference signatures with the first unhashed monitored signatures:
hashing the first unhashed monitored signatures to form corresponding first hashed monitored signatures;
comparing the first hashed monitored signatures with a library of reference hashed signatures to identify the first media event associated with the second time segment of the monitoring data; and
generating media identification data based on the first media event.
10. The method of claim 9 , further including, when the first media event associated with the monitoring data is identifiable based on the first unhashed reference signatures with the first unhashed monitored signatures, not hashing the first unhashed monitored signatures.
11. The method of claim 9 , further including:
determining the watermark coverage based of the first time segment based on a presence of a watermark in the monitoring data; and
comparing the watermark with a reference watermark to determine a second media event associated with the first time segment of the monitoring data.
12. The method of claim 11 , wherein the selection of the first unhashed reference signatures is further based on an offset between a first timestamp associated with the watermark and a second timestamp of the first unhashed reference signatures.
13. The method of claim 11 , wherein the selection of the first unhashed reference signatures is further based on a media source associated with the reference watermark.
14. The method of claim 9 , further including:
selecting second unhashed reference signatures to evaluate for a third time segment of monitoring data based on the comparison of the first unhashed reference signatures with first unhashed monitored signatures from the second time segment of the monitoring data; and
comparing the second unhashed reference signatures with second unhashed monitored signatures from the second time segment of the monitoring data to identify a third media event associated with the third time segment.
15. The method of claim 9 , further including:
selecting third unhashed reference signatures to evaluate for a fourth time segment of monitoring data based on the comparison of the first hashed monitored signatures with a library of reference hashed signatures; and
comparing the third unhashed reference signatures with third unhashed monitored signatures from the fourth time segment of the monitoring data to identify a fourth media event associated with the fourth time segment.
16. A non-transitory computer readable storage medium comprising instructions which, when executed, cause a machine to at least:
select, from a database, first unhashed reference signatures to evaluate for a second time segment of monitoring data based on watermark coverage of a first time segment of the monitoring data, the monitoring data obtained from a meter monitoring media presented by a media device;
compare the first unhashed reference signatures with first unhashed monitored signatures from the second time segment of the monitoring data; and
when a first media event associated with the monitoring data is not identifiable based on the comparison of the first unhashed reference signatures with the first unhashed monitored signatures:
hash the first unhashed monitored signatures to form corresponding first hashed monitored signatures;
compare the first hashed monitored signatures with a library of reference hashed signatures to determine the first media event associated with the second time segment of the monitoring data; and
generate media identification data based on the first media event.
17. The non-transitory computer readable storage medium of claim 16 , wherein the instructions cause the machine to, when the first media event associated with the monitoring data is identifiable based on the comparison of the first unhashed reference signatures with the first unhashed monitored signatures, not hash the first unhashed monitored signatures.
18. The non-transitory computer readable storage medium of claim 16 , wherein the instructions cause the machine to:
determine the watermark coverage based of the first time segment based on a presence of a watermark in the monitoring data; and
compare the watermark with a reference watermark to determine a second media event associated with the first time segment of the monitoring data.
19. The non-transitory computer readable storage medium of claim 18 , wherein the selection of the first unhashed reference signatures is further based on an offset between a first timestamp associated with the watermark and a second timestamp of the first unhashed reference signatures.
20. The non-transitory computer readable storage medium of claim 18 , wherein the selection of the first unhashed reference signatures is further based on a media source associated with the reference watermark.