Abstract:
An incident detection system is proposed based on three incident features:
traffic flow, vehicle angle and vehicle acceleration. The first step after image sequences
acquisition from the video image of CCD camera is vehicle detection, and then
the incident features such as direction of the moving vehicles, traffic flow and the
acceleration are extracted in order to achieve the detection results. This process gives
the best result by total 97.8% of correct rate and 1.02% of false alarm rate within 30 s.
The result shows that this method has a good effect.
提出一种基于事件特征来检测交通事件的监测系统。系统选取 3 个特
征值,交通流、车辆角度和车辆加速度。从 CCD 摄像头获取图像序列之后做车
辆检测,提取车辆移动方向、交通流和车辆加速度来获得检测结果。该方法通
过计算得到正确率 97.8%,误报率 1.02%,检测时间只要 30s。结果表明,该方法
比检测效果更好。