Standoff video analysis for the detection of security anomalies in vehicles

Video surveillance systems are commonly used by security personnel to monitor and record activity in buildings, public gatherings, busy roads, and parking lots. These systems allow many cameras to be observed by a small number of trained human operators but suffer from potential operator fatigue and lack of attention due to the large amount of information provided by cameras which can distract the operator from focusing on important events. In this paper, we propose the design of an autonomous video surveillance system which can operate from a standoff range that analyzes approaching vehicles in order to detect security anomalies. Such anomalies, based on dynamic analysis of the vehicle tracks, include unexpected slowing/stopping or sudden acceleration, particularly near check points or critical structures (e.g. government buildings). A human supervisor can be alerted whenever a significant event is detected and can then determine if the vehicle should be further inspected. Besides dynamic analysis, the system also estimates physical information about the vehicles such as make, body type and tire size. We describe low-complexity techniques to obtain the above information from two cameras.

[1]  David J. Hand,et al.  The Data Sets , 1994 .

[2]  Edward J. Delp,et al.  New models for real-time tracking using particle filtering , 2009, Electronic Imaging.

[3]  Ramakant Nevatia,et al.  A model-based vehicle segmentation method for tracking , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[4]  Edward J. Delp,et al.  Object tracking initialization using automatic moving object detection , 2010, Electronic Imaging.

[5]  Vladimir Petrovic,et al.  Vehicle type recognition with match refinement , 2004, ICPR 2004.

[6]  Nelson H. C. Yung,et al.  Vehicle-type identification through automated virtual loop assignment and block-based direction biased motion estimation , 1999, Proceedings 199 IEEE/IEEJ/JSAI International Conference on Intelligent Transportation Systems (Cat. No.99TH8383).

[7]  David Beymer,et al.  A real-time computer vision system for vehicle tracking and traffic surveillance , 1998 .

[8]  Martin Kampel,et al.  Visual Surveillance for Airport Monitoring Applications , 2006 .

[9]  L. Grammatikopoulos,et al.  AUTOMATIC ESTIMATION OF VEHICLE SPEED FROM UNCALIBRATED VIDEO SEQUENCES , 2005 .

[10]  Mehrübe Mehrübeoglu,et al.  Determination of vehicle speed in traffic video , 2009, Electronic Imaging.

[11]  W. Eric L. Grimson,et al.  Learning Patterns of Activity Using Real-Time Tracking , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  Naoya Ohta,et al.  A statistical approach to background subtraction for surveillance systems , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[13]  Bernhard P. Wrobel,et al.  Multiple View Geometry in Computer Vision , 2001 .

[14]  Nikos Paragios,et al.  Motion-based background subtraction using adaptive kernel density estimation , 2004, CVPR 2004.

[15]  Neil J. Gordon,et al.  A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking , 2002, IEEE Trans. Signal Process..

[16]  David J. Kriegman,et al.  Video-based Car Surveillance: License Plate, Make, and Model Recognition , 2005 .

[17]  Sinem Coleri,et al.  Traffic Measurement and Vehicle Classification with a Single Magnetic Sensor , 2004 .

[18]  Eran A. Edirisinghe,et al.  Two-dimensional statistical linear discriminant analysis for real-time robust vehicle-type recognition , 2007, Electronic Imaging.