System and method for semantic video content analysis
View Patent ↗Embodiments of the invention may include generating a description of a monitored scene and creating a plurality of rules related to the monitored scene by using common language. The description of the monitored scene and the plurality of rules may be converted into system parameters and an analysis of the video data may be performed based on the system parameters. The results of the analysis may be converted into one or more output terms from the plurality of predefined terms and may be displayed to a user in a common language.
1. A method for semantics based video analysis, the method comprising:
accepting a textual description of a monitored scene, the textual description comprising one or more textual terms selected by a user from a plurality of predefined textual terms;
accepting a plurality of rules related to the monitored scene, the rules comprising one or more textual terms from the plurality of predefined textual terms;
converting the textual description of the monitored scene into video characterization parameters;
receiving video data from a plurality of cameras located at the monitored scene;
performing video analysis of the video data based on the video characterization parameters and the rules;
converting results of the video analysis into one or more alerts selected from a plurality of predefined alerts; and
issuing the one or more alerts.
2. The method of claim 1 , further comprising:
receiving an image of the monitored scene divided into a plurality of segments.
3. The method of claim 2 , wherein the plurality of segments are labeled by the one or more textual terms selected by the user to describe the monitored scene.
4. The method of claim 1 , wherein the plurality of rules related to the monitored scene comprises a plurality of events to be detected.
5. The method of claim 1 , wherein performing the analysis of the video data comprises executing a plurality of computer vision algorithms.
6. The method of claim 4 , wherein the analysis of the video data comprises identifying the events.
7. The method of claim 1 , further comprising:
converting the plurality of rules into video characterization parameters.
8. The method of claim 2 , further comprising:
identifying characteristics of the plurality of segments and refining the plurality of segments based on the characteristics.
9. The method of claim 8 , wherein identifying characteristics of the plurality of segments of a first video channel is used to generate a description of a monitored scene of a second video channel.
10. The method of claim 1 , further comprising:
identifying characteristics of the plurality of rules and automatically suggest other rules based on the characteristics.
11. A video analytics system comprising:
a workstation to accept a textual description of a monitored scene from a user and a plurality of rules related to the scene, the textual description and the rules comprising one or more textual terms selected by a user from a plurality of predefined textual terms;
a computing device to convert the textual description of the monitored scene into video characterization parameters; and
a data server to save video data recorded by a plurality of cameras, to receive the video characterization parameters and to perform an analysis of the video data based on the video characterization parameters and the rules;
wherein the computing device is to convert results of the analysis into one or more alerts selected from a plurality of predefined alerts to be presented to the user by the workstation.
12. The system of claim 11 , wherein the computing device is to receive an image of the monitored scene that comprises a plurality of segments of the monitored scene.
13. The system of claim 12 , wherein each of the plurality of segments is labeled with the one or more textual terms selected by the user to describe the monitored scene.
14. The system of claim 11 , wherein the plurality of rules related to the monitored scene comprises a plurality of events to be detected.
15. The system of claim 11 , wherein the analysis of the video data comprises execution of a plurality of computer vision algorithms.
16. The system of claim 14 , wherein the analysis of the video data comprises identifying the events.
17. The system of claim 11 , wherein the computing device is to convert the plurality of rules into video characterization parameters.
18. An article comprising a non-transitory computer-readable storage medium, having stored thereon instructions, that when executed on a computer, cause the computer to:
accept a textual description of a monitored scene, the description comprising one or more textual terms selected by a user from a plurality of predefined textual terms;
accept a plurality of rules related to the monitored scene, the rules comprising one or more textual terms from the plurality of predefined textual terms;
convert the textual description of the monitored scene into video characterization parameters;
receive video data from a plurality of cameras located at the monitored scene;
perform video analysis of the video data based on the video characterization parameters and the rules;
convert results of the video analysis into one or more alerts selected from a plurality of predefined alerts; and
issue the one or more alerts.