Methods and apparatus for identifying audio/video content using temporal signal characteristics

• US 20060195861A1
• Filed: 04/14/2006
• Published: 08/31/2006
• Est. Priority Date: 10/17/2003
• Status: Active Grant

• Current Assignee: The Nielsen Company LLC (MOOD Media Corporation)
• Original Assignee: The Nielsen Company LLC (MOOD Media Corporation)
• Inventors: Lee, Morris
• Granted Patent: US 7,650,616 B2
• Time in Patent Office: Days
• US Class Current: 725/19
• CPC Class Codes: G06F 2218/08 Feature extraction
• CPC Class Codes: G06V 20/40 in video content extracting overlay text G06V20/62; video retrieval G06F16/70; processing of video elementary streams in video servers H04N21/234; processing of video elementary streams in video clients H04N21/44
• CPC Class Codes: H04H 60/35 Arrangements for identifying or recognising characteristics with a direct linkage to broadcast information or to broadcast space-time, e.g. for identifying broadcast stations or for identifying users
• CPC Class Codes: H04H 60/58 of audio determination or detection of speech characteristics in general G10L25/00; speech recognition in general G10L15/00
• CPC Class Codes: H04H 60/59 of video recognising characters or patterns in general G06F18/00, G06V20/00
• CPC Class Codes: H04N 21/4394 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 Class Codes: H04N 21/44008 involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream arrangements characterised by components specially adapted for monitoring, identification or recognition of video in broadcast systems H04H60/59
• CPC Class Codes: H04N 21/8352 involving content or source identification data, e.g. Unique Material Identifier [UMID]

Abstract

Methods and apparatus for identifying audio/video content using temporal characteristics of a signal are disclosed. The disclosed apparatus and methods detect a plurality of signal features of a media signal derived from audio/video content and calculate time intervals between a plurality of those signal features to generate signatures. The signatures are then used to identify the audio/video content.

Claims

1-83. -83. (canceled)

84. A method of identifying audio/video content, comprising:

calculating index values based on elements of a signature associated with the audio/video content;

querying a database based on the index values to form query results; and

identifying the audio/video content based on a statistical analysis of the query results.

85. A method as defined in claim 84, wherein the index values are hash keys.

86. A method as defined in claim 84, wherein the elements of the signature comprise a plurality of N-tuples.

87. A method as defined in claim 86, wherein calculating the index values comprises determining a hash value associated with the elements in an N-tuple.

88. A method as defined in claim 84, wherein the database comprises a hash table.

89. A method as defined in claim 84, further comprising adjusting values associated with the query results based on a time difference between time stamp data associated with a first one of the elements of the signature and time stamp data associated with a second one of the elements of the signature.

90. A method as defined in claim 89, further comprising offsetting the time stamp data associated with the query results based on the time difference.

91. A method as defined in claim 84, wherein identifying the audio/video content based on the statistical analysis of the query results comprises:

generating a histogram of the query results;

identifying a most frequently occurring search result; and

comparing a frequency of the most frequently occurring search result to a predetermined threshold.

92. A method as defined in claim 84, further comprising offsetting at least one of the elements of the signature before calculating the index values.

93. An apparatus for identifying audio/video content, comprising:

a memory; and

a processor coupled to the memory and programmed to;

calculate index values based on elements of a signature associated with the audio/video content;

query a database based on the index values; and

identify the audio/video content based on a statistical analysis of query results.

94. An apparatus as defined in claim 93, wherein the elements of a signature associated with the audio/video content comprise a plurality of N-tuples.

95. An apparatus as defined in claim 93, wherein the processor is programmed to calculate index values by determining a hash value associated with the elements in an N-tuple associated with each index value, wherein the N-tuple comprises a plurality of time values.

96. An apparatus as defined in claim 95, wherein the database comprises a hash table.

97. An apparatus as defined in claim 93, wherein the processor is programmed to identify the audio/video content based on the statistical analysis of query results by:

generating a histogram of the query results;

identifying a most frequently occurring query result; and

comparing a frequency of the most frequently occurring query result to a predetermined threshold.

98. A machine readable medium having instructions stored thereon that, when executed, cause a machine to:

calculate index values based on elements of a signature associated with the audio/video content;

query a database based on the index values; and

identify the audio/video content based on a statistical analysis of query results.

99. A machine readable medium as defined in claim 98 having instructions stored thereon that, when executed, cause a machine to:

generate a histogram of the query results;

identify a most frequently occurring search result; and

compare a frequency of the most frequently occurring search result to a predetermined threshold.

100. A machine readable medium as defined in claim 98 having instructions stored thereon that, when executed, cause a machine to offset time stamp data associated with the query results based on a time difference between time stamp data associated with a first element of the signature and time stamp data associated with a second element of the signature.

101. A machine readable medium as defined in claim 98 having instructions stored thereon that, when executed, cause a machine to offset an element of the signature before calculating the index value.