Tracking of Traffic Monitoring Targets in Complicated Traffic Scene Based on MeanShift Algorithm

To avoid the the interference of busy backgrounds when tracking, detecting and recognizing moving targets in complicated traffic scene, an improved algorithm is proposed on the basis of the original MeanShift algorithm which use different colors of the centroid positions to identify the target. MeanShift algorithm can be used to calcucte the centroid position of each color in the monitoring area. Then the centroid positon of every color in every frame can be identified by analyzing spatial distribution and iteration. At last, establish weighting functions to increase the recognition accuracy so as to recognize and track the targets in complicated traffic scene. Experiments have shown that the improved algorithm is better than the traditional algorithm in identifying and tracking moving targets in the monitoring of complicated traffic scene.