Architecture for Cluster-Based Automated Surveillance Network for Detecting and Tracking Multiple Persons

The demand for surveillance systems to detect, identify, and track objects of interest across large areas calls for scalable camera networks with local tracking decisions enabling efficient feature extraction and reporting. This paper describes one such smart system, comprised of a cluster-based architecture employing hierarchical algorithms to aptly fuse features for detection and tracking of several persons simultaneously. In the proposed architecture, cameras of similar fields of view and proximities form a cluster; clusters act as elements in the multi-camera network, enabling cluster-to-cluster surveillance handoff as opposed to the traditional camera-to-camera handoff of tracked persons, thus greatly decreasing the network bandwidth utilization and complexity. The paper assumes the network has configured itself a priori into clusters, and is aware of adjacent clusters. For each tracked person, one camera per cluster is selected with the 'best' view and acts as the camera selection manager (CSM) for the person. The system adaptively selects the CSM via a three-phase algorithm to (1) extract features of interest based on tracking, (2) fuse these features to compute a robust camera selection score, and (3) broadcast the camera selection scores to the current CSM to select the next CSM. A Kalman filter is used for tracking the features of interest in a cluster. This paper proposes a network-wide (inter-cluster) architecture for detecting and tracking multiple persons, while the paper focuses on intra-cluster processing of a single person. Experimental results for intra-cluster tracking and simulation results for network-wide (inter-cluster) tracking of multiple persons are provided.

[1]  Andrew Gilbert,et al.  Tracking Objects Across Cameras by Incrementally Learning Inter-camera Colour Calibration and Patterns of Activity , 2006, ECCV.

[2]  Mubarak Shah,et al.  A framework for intelligent sensor network with video camera for structural health monitoring of bridges , 2005, Third IEEE International Conference on Pervasive Computing and Communications Workshops.

[3]  Mario Gerla,et al.  Adaptive Clustering for Mobile Wireless Networks , 1997, IEEE J. Sel. Areas Commun..

[4]  Paul A. Viola,et al.  Robust Real-Time Face Detection , 2001, International Journal of Computer Vision.

[5]  Huang Lee,et al.  Collaborative node localization in surveillance networks using opportunistic target observations , 2006, VSSN '06.

[6]  Hideki Hashimoto,et al.  Cluster-camera networking and geometric configuration for intelligent space , 2003, Proceedings 2003 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM 2003).

[7]  Mohan M. Trivedi,et al.  Intelligent environments and active camera networks , 2000, Smc 2000 conference proceedings. 2000 ieee international conference on systems, man and cybernetics. 'cybernetics evolving to systems, humans, organizations, and their complex interactions' (cat. no.0.

[8]  S. Haykin,et al.  Adaptive Filter Theory , 1986 .

[9]  Paul A. Viola,et al.  Robust Real-time Object Detection , 2001 .

[10]  Pramod K. Varshney,et al.  Automatic camera selection and fusion for outdoor surveillance under changing weather conditions , 2003, Proceedings of the IEEE Conference on Advanced Video and Signal Based Surveillance, 2003..

[11]  A. Kak,et al.  A Look-up Table Based Approach for Solving the Camera Selection Problem in Large Camera Networks , 2006 .

[12]  Mohan M. Trivedi,et al.  Networked omnivision arrays for intelligent environment , 2001, SPIE Optics + Photonics.