In the application of computer vision technique, target tracking in image sequences was an important research subject. This paper describes the particle filter and introduces a tracking algorithm based on Kalman particle filter. The algorithm improves the traditional particle filter, whose non-linear and non-Gaussian may result in non-robustness of tracking process. Kalman particle filter use kalman filter to predict the particle’s state and generate the proposal distribution, the state of each particle evolved by the Kalman prediction equations and update equations, increasing the robustness of tracking. Experimental results show that the proposed method in comparison with the traditional particle filtering can be more accurate on tracking and ensure the robustness of performance in a complex environment.