A compactmodular active vision system formulti-target surveillance

This paper presents an omnidirectional active vision system that has been developed for the autonomous acquisition of detailed images of multiple targets. Omnidirectional and perspective camera technologies are integrated to create a robust vision system that combines the strengths of both camera types. A compact, inexpensive and highly modular design is presented in which system modules are stacked vertically. The vertical structure provides each module with an unobstructed 360 degree horizontal view of the surroundings and allows the omnidirectional cameras to directly guide an active camera to view a target point. The physical system design is detailed, along with a description of the system's hardware and software architectures. The hardware architecture is scalable and fully self contained, while the software architecture is built around a user datagram protocol (UDP) network, allowing the computational load to be distributed over multiple computers.

[1]  C. Diehl,et al.  Scheduling an active camera to observe people , 2004, VSSN '04.

[2]  Michael D. Naish,et al.  Scheduling active camera resources for multiple moving targets , 2009, 2009 Canadian Conference on Electrical and Computer Engineering.

[3]  L. Marcenaro,et al.  Dual camera intelligent sensor for high definition 360 degrees surveillance , 2005 .

[4]  Larry S. Davis,et al.  Constructing task visibility intervals for video surveillance , 2006, Multimedia Systems.

[5]  Horst-Michael Groß,et al.  A multi-modal system for tracking and analyzing faces on a mobile robot , 2004, Robotics Auton. Syst..

[6]  Heinrich Niemann,et al.  Statistical modeling and performance characterization of a real-time dual camera surveillance system , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[7]  Motonori Doi,et al.  Real-time video surveillance system using omni-directional image sensor and controllable camera , 2003, IS&T/SPIE Electronic Imaging.

[8]  Helder Araújo,et al.  A General Framework for the Selection of World Coordinate Systems in Perspective and Catadioptric Imaging Applications , 2004, International Journal of Computer Vision.

[9]  Edward M. Riseman,et al.  A Fault-Tolerant Distributed Vision System Architecture for Object Tracking in a Smart Room , 2001, ICVS.

[10]  Alex M. Andrew,et al.  2D Object Detection and Recognition: Models, Algorithms and Networks , 2003 .

[11]  Yasushi Yagi,et al.  Map generation for multiple image sensing sensor MISS under unknown robot egomotion , 1997, Proceedings of the 1997 IEEE/RSJ International Conference on Intelligent Robot and Systems. Innovative Robotics for Real-World Applications. IROS '97.

[12]  Zhengyou Zhang,et al.  Token tracking in a cluttered scene , 1994, Image Vis. Comput..

[13]  Michael D. Naish,et al.  Developing a Modular Active Spherical Vision System , 2005, Proceedings of the 2005 IEEE International Conference on Robotics and Automation.