车头图像采集数量统计的拥堵情况判断仿真

Simulation of Congestion Judgment Based on Image Acquisition Quantity Statistics of Car Front

智能化的车辆数目判断对拥堵治理有着重要的意义。不同时段交通路段的车流状况不同,智能图像拥堵识别多是通过对车头数量的统计完成是否拥堵的判断,但是,拥堵状态下,车头图像往往存在高度的重叠,传统的识别方法在进行分割过程中,受到重叠干扰,很容易出现较大的误差,对拥堵的判断结果不准。提出了一种依据车头图像采集数量统计的拥堵情况判断方法,分析了车头图像采集、灰度化以及二值化处理过程,通过滤波方法去除车头图像中的噪声,采用腐蚀方法对车头图像进行形态学处理,使用bwlable函数对被腐蚀的车头图像区域进行标记,统计车头图像数量,采用模糊综合方法依据车头图像数量准确判别交通拥堵情况。实验结果表明,利用所提方法可对道路拥堵状况进行准确判断,判断的精度高于传统方法。

Intelligent judgment for the number of vehicles has important significance for congestion management. In the paper, a congestion judgment method based on image acquisition quantity statistics of the car fronts was pro- posed. The image acquisition of the car fronts, the gray processing and the binarization processing were analyzed. Through the filter method, the noise in the image was removed. Corrosion method was carried out for morphology pro- cessing on the image of car front, and also Bwlable function was employed to mark the area on the corroded image of car front. The quantity statistics for the images of front cars were carried out, so as to accurately judge traffic conges- tion situation using the fuzzy comprehensive method based on the image quantity of the car fronts. Experimental re- sults show that the proposed method can make an accurate judgment for the road congestion, and the accuracy is higher than traditional methods.