Image Recognition Traffic Patterns for Wireless Multimedia Sensor Networks

Abstract

The objective of this work is to identify some of the traffic characteristics of Wireless Multimedia Sensor Networks (WMSN). Applications such is video surveillance sensor networks make use of new paradigms related with computer vision and image processing techniques. These sensors do not send whole video sequences to the wireless sensor network, but objects of interest detected by the camera. In order to able to design appropriate networking protocols, a better understanding of the traffic characteristics of these multimedia sensors is needed. In this work(1), we analyze the traffic differences between cameras that send whole coded images and those that first process and recognize objects of interest using Object Recognition techniques.