基于数字图像的道路平面线形恢复的研究  

Extracting Horizontal Alignment Based on Digital Image Processing

作　　者：廖若宇[1] 陈雨人[1] 

机构地区：[1]同济大学道路与交通工程教育部重点实验室,上海200092

出　　处：《交通信息与安全》2011年第2期117-120,129,共5页Journal of Transport Information and Safety

摘　　要：在老路改造过程中,需要道路的平面线形资料。基于对数字图像平面图像特性以及噪声分布规律的特点进行分析,运用特征空间聚类法对图像目标区域进行图像分割,提出了一种基于递归循环思想的噪声筛选算法对目标图像进行再处理,达到剔除噪声的效果。对道路的边线进行了标定及距离的测量,配合GPS设备在实践中进行了应用。实践证明方法在满足准确性的同时,也提高了运算的效率。During the redesign of existing roads,information about current horizontal alignment is needed.Through the analysis of digital images and their noise distribution,feature space clustering is used to segment target areas and a recursive algorithm for filtering noise is then used to reprocess the images.The proposed method is used to extract pavement boundary lines and measure their distances,together with the support from GPS.It is found out that this method can not only enhance the accuracy but also improve the efficiency of data acquisition and extraction.

关 键 词：距离测量 边线标定 图像处理 噪声过滤 特征空间聚类 递归 

分 类 号：U445[建筑科学—桥梁与隧道工程]