无人机广域视频的机动车交通参数计算及分析

基于视频的交通参数提取在智能交通系统中具有重要意义.针对无人机视频， 提出了一种基于大范围无人机视频的机动车交通参数提取方法.首先通过建立无人机广 域镜头图像的校正算法去除交叉口视频图像的广角畸变，然后将像素坐标转化为真实坐 标.其次，利用帧间差分法计算行驶车辆的速度、加速度、车头间距等车辆运行参数，将计 算结果与真实车辆及软件提取运行参数进行对比，结果证明该方法具有较高的准确性和 可行性.最后，实例验证该方法能较准确地获得交叉口车辆运行参数，并将其应用于车辆 通过交叉口全过程的速度变化规律分析.

Calculation and Analysis of the Traffic Parameters of Vehicles Based on the Wide-area Drone Video

The video based traffic parameters extraction has great significance in intelligent transportation systems. For the drone video, a new method of the traffic parameter extraction is proposed in this paper. The correction algorithm of the wide- angle lens of the drone is established for eliminating the video images distortion of the intersection. The relation between the image coordinate and the true coordinate is built. Then the frame difference algorithm is used to obtain the traffic parameters of the vehicles, such as the speed, the acceleration and the distance headway. The high accuracy and feasibility of the algorithm is proved by the comparison of the real parameter, the parameter from this paper and that from the software. Finally, an example is given to verify the practicality of this method as well as to analyze the velocity regulation during the vehicles passing the intersection.