基于子特征的交通车辆检测及跟踪算法

通过对交通场景实际情况的分析，特别是考虑到视频在相邻帧之间的变化，提出了一套鲁棒、实时的背景提取及更新算法．实现了基于背景差分算法的车辆检测和提取。同时，根据车辆的角特征点信息，将卡尔曼滤波器引入车辆跟踪算法中，并借助光学投影方程部分恢复待匹配角特征点的高度信息，以检查匹配算法的准确性。经过大量的单目交通图像序列的测试。表明该算法是鲁棒而且实时的，可以有效地从交通场景中提取出运动车辆，并在视野范围内对其予以跟踪。

Vehicles Detection and Tracking Based on the Sub-Feature

Through analyzing actual tragic scene, especially considering the video change between adjoining frames, this paper presents a robust and real-time method for extracting and updating background, and realizes the algorithm of vehicle detection based on background subtraction. Moreover, according the information of vehicle comer, this paper adopts Kalman filter to track each moving vehicle and uses optic projective transform to partially resume the depth of these comers, which can check the match result. The proposed method has been tested on a number of monocular traftic-image sequences and the experimental results show that the algorithm is robust and real-time, which can effectively extract moving vehicle from traffic scene and track each one within the range of visual field.