In order to solve the problem of tracking failure caused by traditional discriminating kernel correlation filter under similar background

this paper proposes a dual background-aware correlation filter algorithm.The algorithm models the target and the background information according to the background-aware correlation filter algorithm.On this basis

the background information around the target is extracted as a negative sample and added to the learning of the tracker

so as to suppress the background information

increase the distinction between the target and the background

and further improve the accuracy of the algorithm in target detection.Comparison experiment was conducted between the public video sequence OTB100 and 6 current mainstream tracking algorithms.The tracking accuracy and success rate of the proposed algorithm were 84.2 % and 63.4 %respectively

which shows better accuracy compared with other algorithms.The proposed algorithm can deal with the tracking task under the complex environment well and has high robustness.