Aiming at the problems that it is difficult for object tracking algorithm based on correlation filtering to adapt to complex environments such as target deformation and occlusion

a color and scale adaptive object tracking algorithm is proposed. The color space and scale adaptivity are applied to object tracking framework

and the color space of the video sequence is analyzed

then the optimal tracking algorithm is selected to track target through scale adaptivity method. The proposed algorithm is evaluated on 13 benchmark challenging sequences and compared with several main target tracking algorithms. The results show that the proposed approach significantly improves the performance by 8. 48 % and 14. 46 % in tracking precision and overlap precision respectively compared to suboptimal algorithm

discriminative scale space tracker（ DSST）

and average center location error is reduced by 13. 31 pixels. The algorithm can still track target efficiently and accurately when the scale changes and occlusion occur.