A novel tracking strategy which can robustly track an object within a fixed environment was proposed. A robust model-based tracker was defined using kalman filtering combined with recursive least squares (RLS). The tracking was done by fitting successively more elaborate models on the tracked region, and the segmentation was done by extracting the regions of the image that are consistent with the computed model of the target. A competitive and efficient dynamic kalman filter were adopted to adaptively update the object model by adding new stable features as well as deleting inactive features. Then the matching search of the feature object was achieved by RLS. The approach was tested on FIRA Mirosot and tested in the context of ball tracking in the FIRA domain. The results show that the proposed approach can obtain excellent ability of tracking and robustness.

Object tracking,   ,  Extended kalman filter,    , Recursive least squares (RLS) 