Video surveillance system employing video primitives
First Claim
Patent Images
1. A computer-implemented method of video surveillance comprising:
- receiving an input video sequence from a video sensor or a video storage facility;
deriving, using at least one computer processor, at least one first video primitive from said received input video sequence, wherein the at least one first video primitive comprises at least one object descriptor referring to at least one observable attribute of at least one non-stationary object in the video sequence;
detecting at least one stationary object in said received input video sequence without reference to another object in said received input video sequence;
deriving, using at least one computer processor, at least one second video primitive from said received input video sequence, wherein the at least one second video primitive comprises at least one object descriptor referring to at least one observable attribute of the at least one stationary object in the video sequence;
processing at least one rule based on said at least one first video primitive and said at least one second video primitive, said at least one rule is satisfied if the at least one non-stationary object has not been in an exclusion zone around said at least one stationary object for a predetermined length of time, and said at least one rule is not satisfied if the at least one non-stationary object is at least partially within the exclusion zone within the predetermined length of time, said exclusion zone defined around and based on said at least one stationary object, said exclusion zone having a size sufficient to incorporate the at least one stationary object and a portion of the at least one non-stationary object;
generating a response if said at least one rule is satisfied.
7 Assignments
0 Petitions
Accused Products
Abstract
A video surveillance system is set up, calibrated, tasked, and operated. The system extracts video primitives and extracts event occurrences from the video primitives using event discriminators. The system can undertake a response, such as an alarm, based on extracted event occurrences.
376 Citations
47 Claims
-
1. A computer-implemented method of video surveillance comprising:
-
receiving an input video sequence from a video sensor or a video storage facility; deriving, using at least one computer processor, at least one first video primitive from said received input video sequence, wherein the at least one first video primitive comprises at least one object descriptor referring to at least one observable attribute of at least one non-stationary object in the video sequence; detecting at least one stationary object in said received input video sequence without reference to another object in said received input video sequence; deriving, using at least one computer processor, at least one second video primitive from said received input video sequence, wherein the at least one second video primitive comprises at least one object descriptor referring to at least one observable attribute of the at least one stationary object in the video sequence; processing at least one rule based on said at least one first video primitive and said at least one second video primitive, said at least one rule is satisfied if the at least one non-stationary object has not been in an exclusion zone around said at least one stationary object for a predetermined length of time, and said at least one rule is not satisfied if the at least one non-stationary object is at least partially within the exclusion zone within the predetermined length of time, said exclusion zone defined around and based on said at least one stationary object, said exclusion zone having a size sufficient to incorporate the at least one stationary object and a portion of the at least one non-stationary object; generating a response if said at least one rule is satisfied. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19)
-
-
20. A computer-implemented method of video surveillance comprising:
-
receiving an input video sequence from a video sensor or a video storage facility; detecting at least one stationary object in said received input video sequence without reference to another object in said received input video sequence; determining, using at least one computer processor, an exclusion zone around said at least one stationary object, said exclusion zone defined around and based on said at least one stationary object; monitoring said exclusion zone for no object of interest inside said exclusion zone, said exclusion zone having a size sufficient to incorporate the at least one stationary object and a portion of an object of interest; identifying a period of time wherein no object of interest has been inside said exclusion zone; comparing said period of time to at least one rule, said at least one rule is satisfied if said period of time wherein no object of interest has been inside said exclusion zone is greater than a predetermined length of time, and said at least one rule is not satisfied if at least one object of interest is at least partially within the exclusion zone within the predetermined length of time; and generating a response if said period of time is greater than said predetermined length of time. - View Dependent Claims (21, 22)
-
-
23. A video surveillance system comprising:
-
at least one video sensor or a video storage facility providing a video sequence; and a video analysis subsystem, said video analysis subsystem to; receive a video sequence; detect at least one stationary object in said received video sequence without reference to another object in said received video sequence; process at least one rule based on said at least one stationary object, said at least one rule is satisfied if a non-stationary object has not been in an exclusion zone around said at least one stationary object for a predetermined length of time, and said at least one rule is not satisfied if at least one non-stationary object is at least partially within the exclusion zone within the predetermined length of time, said exclusion zone defined around and based on said at least one stationary object, said exclusion zone having a size sufficient to incorporate the at least one stationary object and a portion of the non-stationary object; and generate a response if said at least one rule is satisfied. - View Dependent Claims (24, 25)
-
-
26. A computer-implemented method of video surveillance comprising:
-
receiving an input video sequence from a video sensor or a video storage facility; deriving using at least one computer processor, at least one first video primitive from said received input video sequence, wherein the at least one first video primitive comprises at least one object descriptor referring to at least one observable attribute of at least one non-stationary object in the video sequence; detecting at least one stationary object in the received input video sequence without reference to another object in said received input video sequence; deriving using at least one computer processor, at least one second video primitive from said received input video sequence, wherein the at least one second video primitive comprises at least one object descriptor referring to at least one observable attribute of at least one stationary object in the video sequence; processing at least one rule based on said at least one first video primitive and said at least one second video primitive, said at least one rule is satisfied if the at least one non-stationary object has not been in an exclusion zone around said at least one stationary object for a predetermined length of time, and said at least one rule is not satisfied if the at least one non-stationary object is at least partially within the exclusion zone within the predetermined length of time, said exclusion zone defined with respect to said at least one stationary object, said exclusion zone defined after the at least one stationary object is detected; and generating a response if said at least one rule is satisfied. - View Dependent Claims (27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43)
-
-
44. A computer-implemented method of video surveillance comprising:
-
receiving an input video sequence from a video sensor or a video storage facility; detecting at least one stationary object in said received input video sequence without reference to another object in said received video input video sequence; determining, using at least one computer processor, an exclusion zone around said at least one stationary object after the at least one stationary object is detected, said exclusion zone defined with respect to said at least one stationary object; monitoring said exclusion zone for no object of interest inside said exclusion zone; identifying a period of time wherein no object of interest has been inside said exclusion zone; comparing said period of time to at least one rule, said at least one rule is satisfied to if said period of time wherein no object of interest has been inside said exclusion zone is greater than a predetermined length of time, and said at least one rule is not satisfied if an object of interest is at least partially within the exclusion zone within the predetermined length of time; and generating a response if said period of time is greater than said predetermined length of time. - View Dependent Claims (45)
-
-
46. A video surveillance system comprising:
-
at least one video sensor or a video storage facility providing a video sequence; and a video analysis subsystem, said video analysis subsystem to; receive said video sequence; detect at least one stationary object in said received video sequence without reference to another object in said received video sequence; process at least one rule based on said at least one stationary object, said at least one rule is satisfied if a non-stationary object has not been in an exclusion zone around said at least one stationary object for a predetermined length of time, and said at least one rule is not satisfied if at least one non-stationary object is at least partially within the exclusion zone within the predetermined length of time, said exclusion zone defined after the at least one stationary object is detected; and generate a response if said at least one rule is satisfied. - View Dependent Claims (47)
-
Specification