Performance comparison of Background Estimation algorithms for detecting moving vehicle (original) (raw)

Background subtraction is the one of the crucial step in detecting the moving object. Many techniques were proposed for detected moving object however there are few comparative studies carried out to verify their performance. In this paper a performance comparison of different background subtraction algorithms is carried out from the literature as well as through implementation. We investigate some of the techniques which varying from simple techniques such as frame differencing and approximation median filter, to more complicated probabilistic modeling techniques. Our results show that simple techniques such as approximation median filter can produce good results with much lower computational complexity.