Sparsity-driven bandwidth-efficient decentralized tracking in visual sensor networks

Coşar, Serhan and Çetin, Müjdat (2015) Sparsity-driven bandwidth-efficient decentralized tracking in visual sensor networks. Computer Vision and Image Understanding, 139 . pp. 40-58. ISSN 1077-3142 (Print) 1090-235X (Online)

This is the latest version of this item.

[thumbnail of cosar_CVIU15.pdf] PDF
Restricted to Registered users only

Download (6MB) | Request a copy


Recent developments in low-cost CMOS cameras have created the opportu- nity of bringing imaging capabilities to sensor networks and a new eld called visual sensor networks (VSNs) has emerged. VSNs consist of image sensors, embedded processors, and wireless transceivers which are powered by batter- ies. Since energy and bandwidth resources are limited, setting up a tracking system in VSNs is a challenging problem. In this paper, we present a frame- work for human tracking in VSN environments. The traditional approach of sending compressed images to a central node has certain disadvantages such as decreasing the performance of further processing (i.e., tracking) because of low quality images. Instead, in our decentralized tracking framework, each camera node performs feature extraction and obtains likelihood functions. We propose a sparsity-driven method that can obtain bandwidth-efficient representation of likelihoods extracted by the camera nodes. Our approach involves the design of special overcomplete dictionaries that match the structure of the likelihoods and the transmission of likelihood information in the network through sparse representation in such dictionaries. We have applied our method for indoor and outdoor people tracking scenarios and have shown that it can provide major savings in communication bandwidth without sig- nificant degradation in tracking performance. We have compared the tracking results and communication loads with a block-based likelihood compression scheme, a decentralized tracking method and a distributed tracking method. Experimental results show that our sparse representation framework is an effective approach that can be used together with any probabilistic tracker in VSNs.
Item Type: Article
Uncontrolled Keywords: Camera networks, visual sensor networks, human tracking, sparse representation, designing overcomplete dictionaries
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions: Faculty of Engineering and Natural Sciences > Academic programs > Electronics
Faculty of Engineering and Natural Sciences
Depositing User: Müjdat Çetin
Date Deposited: 24 Dec 2015 16:55
Last Modified: 23 Aug 2019 15:37

Available Versions of this Item

Actions (login required)

View Item
View Item