A Novel Algorithm of Fast Detection and Segmentation of Moving Object
Zhuoyi Lv, Kebin Jia · 2008
Detecting and segmenting moving object is an important subject in computer visual analysis. Firstly, this paper discusses algorithms of detecting moving target from static background in video sequences. Secondly, as the existing mixture Gaussian models can't solve the problems such as ghost and real-time application, a new method based on moving edge detection of difference between adjacent frames is proposed. Experimental results show that the proposed algorithm can get exact moving edge from complex background and eliminate disturbing of background efficiently, with operating time reducing dramatically. Finally, by filling the edge map the target foreground object can be segregated completely. Combined with mathematical morphology, it can reduce noise and smooth out gaps.