Abstract Segmentation of motion image sequences is an important problem in computer vision. In this paper, maximizer of the posterior marginalsmaximum a posteriori (MPMMAP), is adopted based on Bayesian frame for motion segmentation. Firstly, the smoothness term of likelihood function in Bayesian frame is redefined. The region shrinking algorithm is used to estimate the supporting regions of moving objects during the iteration. Then a model is proposed which represents the affine motion with 6 parameters by the center and main axes of a region. Motion parameters are estimated merely by pixels on main axes and derived more quickly than before. The estimation is transformed into a kind of optimal problem with parameters in limited ranges, and DIRECT algorithm is used to compute the motion parameters. Compared with the traditional algorithms, the proposed method improves the accuracy and stability in motion parameter estimation. The results of simulated experiments show the effectiveness of the proposed method.
[1] Zhang Dengsheng. Segmentation of Moving Objects in Image Sequence: A Review. Circuits Systems and Signal Processing, 2001, 20(2): 143183 [2] Liu Danghui, Shen Lansun. Survey on the Segmentation of Moving Object in Video. Journal of Circuits and Systems, 2002, 7(3): 7785,69 (in Chinese) (刘党辉,沈兰荪.视频运动对象分割技术的研究.电路与系统学报, 2002, 7(3): 7785,69) [3] Murry D W, Buxton B F. Scene Segmentation from Visual Motion Using Global Optimization. IEEE Trans on Pattern Analysis and Machine Intelligence, 1987, 9(2): 220228 [4] Bouthemy P, Francois E. Motion Segmentation and Qualitative Dynamic Scene Analysis from an Image Sequence. International Journal of Computing Vision, 1993, 10(2): 157182 [5] Chang M M, Sezan M I, Tekalp A M. An Algorithm for Simultaneous Motion Estimation and Segmentation // Proc of the IEEE International Conference on Acoustics, Speech and Signal Processing. Adelaide, Australia, 1994, V: 221224 [6] Stiller C. ObjectBased Estimation of Dense Motion Fields. IEEE Trans on Image Processing, 1997, 6(2): 234250 [7] Weiss Y. Smoothness in Layers: Motion Segmentation Using Nonparametric Mixture Estimation // Proc of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Juan, Puerto Rico, 1997: 520527 [8] Weiss Y, Adelson E H. A Unified Mixture Framework for Motion Segmentation: Incorporating Spatial Coherence and Estimating the Number of Models // Proc of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Francisco, USA, 1996: 321326 [9] Zhang Wei, Fang Xiangzhong, Yang Xiaokang. Moving Vehicles Segmentation Based on Bayesian Framework for Gaussian Motion Model. Pattern Recognition Letters, 2006, 27(9): 956967 [10] Calderon F, Marroquin J L, Botello S, et al. The MPMMAP Algorithm for Motion Segmentation. Computer Vision and Image Understanding , 2004, 95(2): 165183 [11] Gablonsky J M. An Implementation of the DIRECT Algorithm. Technical Report, CRSCTR9829, Raleigh, USA: North Carolina State University. Center for Research in Scientific Computation, 1998 [12] Bjrkman M, Holmstrom K. Global Optimization Using the DIRECT Algorithm in Matlab. Journal of Advanced Modeling and Optimization, 1999, 1(2): 1727 [13] Gablonsky J M. Modifications of the DIRECT Algorithm. Ph.D Dissertation. Raleigh, USA: North Carolina State University. Department of Mathematics, 2001 [14] Finkel D E, Kelley C T. Convergence Analysis of the DIRECT Algorithm [EB/OL]. [20040714]. http://www.optimizationonline.org/DBHTML/2004/08/934.html
2008, Vol. 21 
