For long-term robust tracking to single target, a corrected tracking algorithm based on static-adaptive appearance model was proposed. Firstly, the interference factors that may be encountered in the tracking process were divided into two categories from environment and target itself, then a static appearance model and an adaptive appearance model were proposed respectively. The static appearance model was used for global matching while the adaptive appearance model was employed for local tracking, and the former corrected tracking drift of the latter. A single-link hierarchical clustering algorithm was used to remove the noise introduced by the fusion of the above two models. To capture the re-occurring target, static appearance model was applied for global search. Experimental results on standard video sequences show that the accuracy of tracking the target center is 0.9, and the computer can process 26 frames per second. The proposed tracking algorithm framework can achieve long-term stable tracking with good robustness and real-time performance.

To solve target tracking with target changes under complex background, an adaptive target tracking method that combined graph-based semi-supervised learning method with the particle filter was proposed. It used LLGC (Learning with Local and Global Consistency) algorithm to establish the cost function, and took current status of the candidate as unlabeled samples, then established diagram using all samples as vertex, taking the optimal solution of the cost function as current status, obtaining the target position in current frame. Besides, it used the tracking result to update the labeled samples in real time, so that the algorithm could adapt to the target deformation, partial occlusion and illumination changes. Analysis and experiment show that the proposed method can handle complicated situations like occlusion or similar background interference very well, and achieves target tracking robustly.