Moving vehicle detection based on an improved interframe difference and a Gaussian model

Abstract

For moving vehicle detection, this paper presents an algorithm on the basis of an improved interframe differential algorithm and an improved Gaussian model. Firstly, according to a statistical histogram, an interesting region is extracted. Through a mean algorithm, an initial background model is established. The interesting region is divided into several blocks by a self-adaptive method. Secondly, according to an improved interframe difference algorithm, the interesting region is separated roughly. On the basis of these steps, we utilize an improved Gaussian model to separate the rough results precisely. At last, the results are processed by double-threshold background subtracting. Experimental results show this algorithm can detect moving vehicles rapidly and accurately.