论文部分内容阅读
为有效提高快速路交通事件检测的覆盖率,解决算法误判率较高的问题,本文通过分析交通事件发生时交通流参数在时间维与空间维的变化,分别提出了基于固定检测器的多参数判别算法与基于浮动车的时空二维判别算法.当两个数据源算法同时满足检测条件时,研究以D-S理论为基础,将两个子算法有效地结合,实现事件的综合检测。最后,研究利用北京市快速路上采集的交通事件数据、固定检测器数据和浮动车数据对算法性能进行了检验。结果表明,基于固定检测器的多参数判别算法和基于浮动车的时空二维判别算法的检测率、误判率都达到了较好的效果,可以满足系统应用的需要。基于D-S理论的综合检测算法,具有比其他经典判别算法与两个子算法更低的算法误判率。
In order to effectively improve the detection coverage of the expressway traffic accidents and solve the problem of high false positive rate of the algorithm, this paper presents the multi-dimensional analysis of traffic flow parameters in time and space, Parameter discriminant algorithm and space-time two-dimensional discriminant algorithm based on floating car.When the two data source algorithms satisfy the detection conditions at the same time, the study combines the two sub-algorithms effectively based on DS theory to realize the comprehensive detection of events. Finally, the paper tests the performance of the algorithm by using the traffic accident data collected on Beijing Expressway, fixed detector data and floating car data. The results show that both detection rate and false positive rate of multi-parameter discriminant algorithm based on fixed detector and space-time two-dimensional discriminant algorithm based on floating vehicle have achieved good results and can meet the needs of system application. The comprehensive detection algorithm based on D-S theory has lower false positive rate than other classical discriminant algorithms and two sub-algorithms.