中文摘要：

为提高交通冲突严重程度的判别精度,基于运动单元圆假设与视频检测技术,提出了交叉口两车交通冲突严重程度量化方法。利用视频处理程序与实地采集的视频图像提取圆单元运动学参数,并通过迭代方法进行了冲突检测与冲突时间判定。充分考虑车辆的制动性能,将非完全制动停车距离与瞬时车速的比值作为冲突严重程度判别临界值,以冲突时间与临界值为参数,建立了冲突严重程度量化值,并对其严重程度进行了量化与分级,以北京市某交叉路口为例进行了应用实例分析。分析结果表明：两运动单元驶入交叉口后,二者之间的冲突严重程度量化值由1.32增大为2.45,后又不断减小直至无交通冲突;随着冲突严重程度的增大,驾驶人会采取紧急制动措施使冲突严重程度减小,直至冲突消解。可见该方法可以方便地检测交通冲突,准确量化交叉口两运动单元之间的冲突严重程度。

英文摘要：

In order to improve the precision in traffic conflict severity discrimination, a quantitative method for conflict severity at intersection was proposed based on the circle assumption of motor units and video detection technology. The moving parameters of motor units were obtained by using the video processing program and the real-time collected video images. Conflict detection and conflict time were determined by using iterative method. Considering the braking performance of vehicle, the critical value of conflict severity was obtained by the ratio of braking distance to instantaneous speed. The quantization value of conflict severity was established taking conflict time and critical value as parameters, and the severity of traffic conflict was quantified and classified. A certain intersection in Beijing was analyzed. Analysis result shows that the quantization values of conflict severity between two motor units increase from 1.32 to 2.45, and then decrease until no conflict exists. Driver would take emergency braking measures with the increasing of conflict severity, and then the conflict severity decreases until disappears. The quantitative result accords with the practical situation, and the method could be used in conflict detection and the severity quantification of traffic conflict conveniently. 3 tabs,8 figs, 16 refs.