中国计量大学学报2017,Vol.28Issue(2):185-189,233,6.DOI:10.3969/j.issn.2096-2835.2017.02.008
墙体表面的网状裂纹最大宽度测量
Measurement of the maximum width of mesh cracks on wall surface
丁宁 1陈晓竹1
作者信息
- 1. 中国计量大学 信息工程学院,浙江 杭州 310018
- 折叠
摘要
Abstract
Detection of concrete wall cracks and measurement of the maximum width of the cracks have attracted much attention since wall cracks could cause severe safety hazard.In this paper, an intelligent crack detection method based on image processing was introduced.According to the distribution of crack pixels, the crack was extracted by using the size of the connected domain, and the impurities such as pseudo-crack were removed.According to the location relation of the pixels of the crack, the initial value of the cluster was obtained, and then the K-means clustering algorithm was used to iteratively calculate the shortest distance between the feature points of the crack and the corresponding line and to classify the pixels in the image as cracks of different directions.The maximum width of the cross crack was obtained by using the categorized crack pixels to perform edge detection and maximum width measurement respectively.The relative error of the maximum crack width was 2.968%, and the relative error of the maximum width of the slanting crack was 5.188%.关键词
裂纹检测/图像处理/K-means聚类算法/特征提取Key words
crack detection/image processing/K-means clustering algorithm/feature extraction分类
信息技术与安全科学引用本文复制引用
丁宁,陈晓竹..墙体表面的网状裂纹最大宽度测量[J].中国计量大学学报,2017,28(2):185-189,233,6.