Author(s): |
Wei He, Department of Computer Science and Engineering, Yunnan University, Kunming, P. R. 650091, China Hui Tang, Department of Computer Science and Engineering, Yunnan University, Kunming, P. R. 650091, China Dan Xu, Department of Computer Science and Engineering, Yunnan University, Kunming, P. R. 650091, China |
Abstract: |
In this work, we develop a new algorithm for detecting of the moving objects in video sequences, which is based on background subtraction method and symmetric difference method. Firstly, we use the Surendra background updating algorithm to model the background and get a primary detection result about the moving objects by background subtraction. In order to adapt to the scene lighting changes in the real time, Ostu algorithm is used to calculate a dynamic threshold in the process of background updating. Secondly, we get the logical OR of the primary detection result and the differential image by the symmetric difference method to obtain the complete moving objects regions. The Experimental results show that this algorithm is rapid, efficient, and fits for real time detection.
|