Intelligent, dynamic, long-term digital surveilance media storage system
First Claim
Patent Images
1. A digital surveillance media storage system, comprising:
- a media analyzer receiving a media stream and decomposing the media stream into a plurality of media segments based on scenes recorded in the media stream, each media segment of the plurality of media segments having an indicia of a starting location for the media segment in the media stream and a duration time for the media segment;
an encoding format selector receiving content semantic information for each media segment of the plurality of media segments and selecting an encoding format for each media segment of the plurality of media segments from plural encoding formats based on the content semantic information for that media segment and scenes recorded in that media segment, wherein the encoding format selector selects;
a low-frame rate or skipping frame encoding when no change in motion is detected;
an object based encoding when low motion objects are detected; and
a frame-based encoding when high motion video is detected;
an encoder engine receiving the plurality of media segments from the media analyzer and encoding the media segments according to the encoding format selected by said encoding format selector, where different encoding formats are selected for different media segments; and
a file writer receiving the plurality of encoded media segments from the encoder engine and storing information describing the plurality of encoded media segments for the media stream in a single media file.
2 Assignments
0 Petitions
Accused Products
Abstract
An Intelligent, dynamic, long-term digital Surveillance Media storage System (ISMS), proposes an intelligent, dynamic, digital media storage mechanism for digital surveillance applications that can analyze media streams in real-time and provide flexible, scalable, and self-adjustable encoding and indexing mechanisms based on actual needs for digital surveillance systems. The system includes several subsystems, such as: (1) a media stream retriever/receiver; (2) a content-based description generator; (3) an intelligent engine with metadata description storage and media file storage (4) an event manager; (5) a media stream writer; and (6) an encoding optimizer.
-
Citations
24 Claims
-
1. A digital surveillance media storage system, comprising:
-
a media analyzer receiving a media stream and decomposing the media stream into a plurality of media segments based on scenes recorded in the media stream, each media segment of the plurality of media segments having an indicia of a starting location for the media segment in the media stream and a duration time for the media segment; an encoding format selector receiving content semantic information for each media segment of the plurality of media segments and selecting an encoding format for each media segment of the plurality of media segments from plural encoding formats based on the content semantic information for that media segment and scenes recorded in that media segment, wherein the encoding format selector selects; a low-frame rate or skipping frame encoding when no change in motion is detected; an object based encoding when low motion objects are detected; and a frame-based encoding when high motion video is detected; an encoder engine receiving the plurality of media segments from the media analyzer and encoding the media segments according to the encoding format selected by said encoding format selector, where different encoding formats are selected for different media segments; and a file writer receiving the plurality of encoded media segments from the encoder engine and storing information describing the plurality of encoded media segments for the media stream in a single media file. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 20)
-
-
18. A digital surveillance media storage system, comprising:
-
a media analyzer receiving a media stream and decomposing the media stream into a plurality of media segments based on scenes recorded in the media stream, each media segment of the plurality of media segments having a duration time associated therewith; an encoding format selector receiving content semantic information for each media segment of the plurality of media segments and selecting an encoding format for each media segment of the plurality of media segments based on the content semantic information for that media segment and scenes recorded in that media segment, wherein the encoding format selector selects a low-frame rate or skipping frame encoding when no change in motion is detected; an object based encoding when low motion objects are detected; and a frame-based encoding when high motion video is detected; an encoder engine receiving the plurality of media segments from the media analyzer and encoding the media segments according to the encoding format selected by said encoding format selector, where different encoding formats are selected for different media segments a file writer receiving the plurality of encoded media segments from the encoder engine and storing information describing the plurality of encoded media segments for the media stream in a single media file; and an encoding optimizer periodically retrieves the media file and selectively re-encodes a subset of the plurality of media segments which comprise the media stream. - View Dependent Claims (19, 21)
-
-
22. A digital surveillance system, comprising:
-
a plurality of surveillance cameras monitoring a surveillance domain and generating a media stream of the surveillance domain; a domain knowledge profile defining settings of the digital surveillance system based on security requirements of the surveillance domain; a media analyzer receiving the media stream and decomposing the media stream into a plurality of media segments based on scenes recorded in the media stream, each media segment of the plurality of media segments having an indicia of a starting location for the media segment in the media stream and a duration time for the media segment; an encoding format selector receiving content semantic information for each media segment of the plurality of media segments and selecting an encoding format for each media segment of the plurality of media segments from plural encoding formats based on a) the content semantic information for that media segment, b) scenes in that media segment, and c) the domain knowledge profile, wherein the encoding format selector selects a low-frame rate or skipping frame encoding when no change in motion is detected; an object based encoding when low motion objects are detected; and a frame-based encoding when high motion video is detected; an encoder engine receiving the plurality of media segments from the media analyzer and encoding the media segments according to the encoding format selected by said encoding format selector, where different encoding formats are selected for different media segments; and a file writer receiving the plurality of encoded media segments from the encoder engine and storing information describing the plurality of encoded media segments for the media stream in a single media file. - View Dependent Claims (23, 24)
-
Specification