Abstract: Aiming to the high-definition traffic video collected by airship flying along road, this paper proposed a vehicle tracking approach based on Monte-Carlo algorithm. The particles state characteristics such as position and scale can be obtained by Monte-Carlo algorithm to forecast. Since the similarity between each forecasted state and the state of reference target can be derived, the state with maximum similarity can be identified, which is treated as the tracking result. The Experiment research show that this approach can track the irregular and nonlinear moving vehicles real-time and accurately, deal with the situation of short time occlusion as well.