基于小波变换的压线检测算法的设计与实现

Design and implementation of vehicle pressing yellow line detection based on wavelet

针对我国交通行业的发展现状,为了研发基于视频的车辆违禁压线检测系统的需要,在对智能交通系统进行分析和研究的基础上,客观地比较和借鉴了现有的视频监控技术,根据交通道路中央黄线区域的分布特点,提出了基于小波变换的压线检测算法,实现了对车辆轧压道路中央黄线区域的检测。首先利用小波变换的图像分割技术,将道路中央的黄线区域分割出来,以缩小检测范围;然后运用基于颜色直方图的图像分割匹配算法判断车辆是否轧压了黄线区域。在此基础上,开发了车辆违禁压线检测原型系统。实验结果表明,该算法在检测成功率、误检率和时效性等方面都优于其它算法,更能满足工程实践的需要。

Accarding to the China＇s transportation status,to develop vehicle pressing yellow line detection system,intelligence transportation system（ITS） is introduced,and a varity of technologies in regard with yellow line detection is discassed briefly,and then a new detection method based on wavelet is proposed,which exploiting yellow line distribution characteristics.This algorithm first segments the yellow line from original image using wavelet decomposition,thus can decrease detection area,then using color histogram based area match algorithm to detect whether there is a vehicle pressing the yellow line.Finally,a proto system is designed.Experimental results show that the algorithm is superior than many others in terms of detection accuracy,false detection percentage and computation complexity,so the system is more practical.