A two-step approach for tracking moving object is proposed in this paper. According of this approach, the structure template of object is firstly extracted with morphologic methods, and then the object tracking is performed with the structure template. The structure template is composed with the stable edges and the cross-points, which can describe the essential structure information of objects. The tracking processing can be divided into two steps: in the first step, the structure template is wholly moved very closed to the tracked object, and in the second step, the structure template is modified to converge to the cross-points and the edges in object image. Because of considering the structure information of object, the robusticity of tracking can be improved greatly.
Kass M, Witkin A, Terzopoulos D. Snakes: Active contour models.Proc. First International Conference on Computer Vision, London,English, 1987:259 - 268.[2]Terzopoulos D, Szeliski R. Tracking with Kalman Snakes in active vision. Cambridge, MA, USA: MIT Press, 1992:3 - 20.[3]Peterfreund N. Robust tracking with Spatio-snakes: Kalman filtering approach. Proc. the 1998 IEEE 6th International Conference on Computer Vision, Bombay, India, 1998:433 - 439.[4]Peterfreund N. The velocity snake: Deformable contour for tracking in spatio-velocity space[J].Computer Vision and Image Understanding.1999, 73(3):346-[5]Yue F, Erdem A T. Tracking visible boundary of objects using occlusionadaptive motion snake[J].IEEE Trans. on Image Processing.2000, 9(12):2051-[6]Yu Z, Jain A K, Dubuisson M P. Object tracking using deformable templates[J].IEEE Trans. on Pattern Analysis and Machine Intelligence.2000, 22(5):544-[7]Bleau A, Leon L J. Watershed-based segmentation and region merging[J].Computer Vision and Image Understanding.2002, 77(3):317-