To solve efficient problem of object tracking in terms of texture changes

illumination changes

rotational changes and even occlusion

propose a real time target tracking algorithm based on correlation filtering of spatial-temporal information. The proposed algorithm has considered that the object is relative to surrounding information in spatial area

which can be used to construct target spatial filtering model in the process of target localization. Moreover

this algorithm introduces temporal variables to update spatial filtering model online

using information similarity between consecutive frames to decrease effects of occlusion on tracking. Through experiment contrast

it is verified that the proposed algorithm has strong robustness and good real-time performance when the target is under interference of non-ideal factors.