Moving object detecting using gradient information, three-frame- differencing and connectivity testing

Abstract

The motivation and the current status of moving object detecting research were reviewed firstly, and a novel method was proposed and verified in this paper. The method works mainly by using gradient information, threeframe- differencing and connectivity-testing-based noise reduction. The results of theoretical analyses and computer simulation show that the method has some advantages over its competitors, e.g., a wider application range, a less computation complexity and a faster processing speed. Thus, it is able to work rather robust in a noisy environment. © Springer-Verlag Berlin Heidelberg 2006.