论文部分内容阅读
基于视频图像处理方法,提出了一种复式伸缩窗来实时跟踪交叉口排队车辆队尾和队头的位置变化,从而准确描述交叉口车辆排队形成和消散过程。通过检测指定区域内车辆是否存在和是否运动,分别构建跟踪排队队尾和队头的队尾伸缩窗和队头伸缩窗。描述排队首尾伸缩变化的复式伸缩窗则由这两个伸缩窗相互协作所构成。根据跟踪队尾和队头的结果,车辆排队长度和停车延迟时间等重要参数就可以轻易得到。试验结果表明本文算法能实时准确地跟踪队尾和队头的位置,能适应不同天气环境和光照变化,其准确率达到92%以上,较好地满足车辆堵塞监控和交通信号灯控制的需要。
Based on the video image processing method, a double retractable window is proposed to track the changes of the position of the tail and the head of queuing vehicles at intersections in real time so as to accurately describe the formation and dissipation process of queuing at the intersections. By detecting the existence and movement of the vehicles in the designated area, the tail telescopic windows and the head telescopic windows which track the tail of the queue and the head of the queue are constructed respectively. The double telescoping window, which describes the change in line-up and contraction, is made up of the two telescopic windows that cooperate with each other. According to tracking the results of the tail and the head of the queue, vehicle queuing length and parking delay time and other important parameters can be easily obtained. The experimental results show that the proposed algorithm can track the position of the tail and the head of the team in real time and can adapt to different weather conditions and illumination changes. The accuracy of the algorithm can reach above 92%, which can well meet the needs of vehicle congestion monitoring and traffic signal control.