交通事件视频图像自动识别系统的研究

如何有效利用高速公路视频图像信息，实时全天候自动智能检测交通事件，提升交通管理部门应急处置能效，是当前公路视频监控亟需解决的问题．本文基于视频图像处理技术，开展交通事件自动识别算法和系统的研究：采用中值滤波、亮度缩放等图像预处理方法，提取交通视频图像的前景目标边界并抑制噪声；基于灰度阈值化方法，对车辆前景进行二值化分割处理；提出一种二值化场景图像连通区域标定算法，对交通事件前景目标进行特征提取与检测识别，并基于上述算法和识别流程开发了交通事件视频自动识别系统．试验表明，该系统对噪声干扰抑制能力较强，识别准确率较高．

Study of traffic incident automatic recognition system based on video image

Presently it is extremely urgent for highway＇s visual monitors to make full use of visual in formation on expressway, detect traffic incidents intelligently 24-hours, and raise first-aid treatment efficiency to the department of traffic management. Based on the camera set on the expressway,video image processing was used to detect the traffic incident. Firstly, in order to extract boundary of fore- ground objects and inhibit noise, the image pre-processing methods of median filter and image scaling were employed. Binarization segmentation of foreground objects was processed based on the threshold value algorithm. And then, a connected region labeling algorithm of binarization scene image was pro- posed to extract feature of foreground objects and detect the traffic incidents. Experimental results show that the algorithm can filter the waves and decline the noisy effectively. The system can recog- nize traffic incidents accurately and rapidly.