Resource-aware distributed particle filtering for cluster-based object tracking in wireless camera networks

Kihyun Hong, Purdue University

Abstract

The proliferation of miniaturized low-power computing devices, advances in wireless communications, and the availability of inexpensive imaging sensors have enabled the development of wireless camera networks (WCN). In this dissertation, we consider the problem of real-time object tracking with a WCN. Existing object tracking methods designed for multi-camera systems do not take into account the unique constraints of WCNs. Specifically, an effective object tracking system for WCNs must anticipate unreliable network communication, limited memory, and limited computational power in each camera node. In particular, unreliable communication degrades the quality of the visual information shared by the cameras, which ultimately degrades the tracking performance in the network. We present a novel resource-aware framework for the implementation of distributed particle filters in resource-constrained WCNs. Our method focuses on the effects of communication failures on object tracking performance by adjusting the amount of data packets generated and transmitted by the cameras according to the network conditions. We demonstrate the performance of the proposed framework using three different mechanisms to share the particle information among nodes: synchronized particles, Gaussian mixture models, and Parzen windows. We show that all three approaches benefit from the proposed resource-aware mechanism in terms of tracking accuracy or energy efficiency.

Degree

Ph.D.

Advisors

Tan, Purdue University.

Subject Area

Computer Engineering|Electrical engineering

COinS