Abstract: In order to solve the problems during the target tracking process such as scale change, occlusion and tracking drift, an integrated tracking and detection algorithm was proposed, to which was added the scale change update mechanism based on the minimum mean square error (MOSSE) of the output, and the latest position and the optimal size of the target was calculated through the relevant calculation. The analysis of the tracking experiments of 51 test video sets on Benchmark demonstrates that the proposed algorithm performs better than traditional algorithm in terms of scale variation and occlusion resistance.