To solve multiple moving objects segmentation problem in complex traffic scene outdoors, a new segmentation algorithm of multiple moving objects segmentation in complex traffic scene of outdoors was developed. The presented method consists of two phases. In the motion detection phase, motion-changed regionwas determined using an improved level set algorithm based on Mumford-Shah model of two consecutive frames difference of image sequence, and an simplified clustering algorithm providing initialization evolution curves of level set algorithm. Motion window was generated through detecting motion-changed region. In themotion segmentation phase, regions segmentation using an improved weighted k-means clustering algorithm of moving windows area were merged based on region motion similarity criterion. Computation time was reduced because of avoiding processing of the whole image. Very promising experimental results indicates that the proposed algorithm is effective with strong robustness using real image sequences of complex traffic scene and can resolve occlusions among multiple moving objects.
�� ��,������,�� ��. �����˶������ɵ�ʱ����Ƶ�ָ�[J] ����ѧ��,2004,32(3):480~484 Zhu Hui,Li Zaiming, Cai yi. Spatio-temporal vido segmentation based on the generation of motion window[J]. Acta Electronica Sinica,2004,32(3):480~484(in Chinese)
Fan J, Yu J, Fu G, %et al%. Spatiotemporal segmentation for compact video representation[J]. Signal Processing,2001,16:553~566
Geidenberg R,Kimmel R,Rivlin E, %et al%. Fast geodesic active contours[J].IEEE Transactions on Image Processing.2001,10(10):1467-1475
Chan F T, Vese L. Active contours without edges[J].IEEE Trans Image Processing.2001,10(2):266-277
�� ��, �� ��, ʩ����. ����Mumford-Shahģ�͵Ŀ���ˮƽ��ͼ��ָ��[J]. �����ѧ��,2002,11(25):1175~1183 Li Jun, Yang Xin,Shi Pengfei. A fast level set approach to image segmentation based on Mumfor-Shah model[J]. Chinese Journal of Computers,2002 , 11(25):1175~1183(in Chinese)
Kottke D, Sun Y. Motion estimation via cluster matching[J].IEEE Transactions on Pattern Analysis and Machine Intelligence.1994,16:1128-1132