Fast Moving Region Detection Scheme in Ad Hoc Sensor Network

In this paper we present a simple yet effective temporal differencing based moving region detection scheme which can be used in limited resource condition such as in ad-hoc sensor network. Our objective is to achieve realtime detection in these low-end image sensor nodes. By double-threshold temporal differencing we can exclude the effect of global motion as well as detect the real motion regions. Then to achieve fast processing speed and overcome foreground aperture problem, we scale down the searching space to a rather small size, 30x40, and apply our Scalable Teeterboard Template to locate moving regions’ bounding boxes. Resource requirement and time complex of our scheme are very low yet experiment result shows that our scheme yields a high detection rate. And our scheme’s speed and detection rate cannot be affected essentially by the number of objects in the field of view.