摘要
为了提高公路混凝土路面病害的检测效率,采用图像处理技术实现了路面裂缝的自动识别和病害参数的自动提取。首先通过阈值分割法检测出路面裂缝,根据线性和网状裂缝图像背景连通域数量的差异,采用区域标记法计算背景连通域的数量实现了裂缝分类;其次根据线性裂缝几何形态的差异性,通过投影法实现了横、纵和斜向裂缝的细分;最后对线性裂缝提取裂缝骨架并细化求出其长度、平均宽度等病害参数;对于网状裂缝通过求其最小外接矩形计算其破损面积。
In order to improve the detecting efficiency of the concrete highway pavement disease,image processing technology is applied to automatically recognize cracks,and extract the disease parameters.The crack can be detected by threshold segmentation methods.According to quantity difference of the background connected regions in the linear and reticular crack image,region labeling method is adopted to calculate quantity of background connected regions to classify cracks.According to the geometric shape differences of three kinds of linear crack,the transverse,longitudinal and diagonal crack can be subdivided by projection method.And disease parameters such as length and mean width of linear crack are measured after its skeleton is extracted by thinning.The damaged area in the reticular crack image is calculated by finding its minimum enclosing rectangle.
出处
《传感器与微系统》
CSCD
北大核心
2013年第4期61-64,共4页
Transducer and Microsystem Technologies
基金
陕西省教育厅自然专项科研计划资助项目(09JK483)
关键词
混凝土路面
裂缝检测
图像处理
concrete pavement
crack detection
image processing
