基于图像处理的混凝土表面蜂窝麻面检测  

Flaw Detection of Concrete Surface Based on Image Processing

作　　者：栾星 LUAN Xing(China Telecom Co.,Ltd.,Hubei Branch,Wuhan,Hubei Province,430000 China)

机构地区：[1]中国电信股份有限公司湖北分公司,湖北武汉430000

出　　处：《科技创新导报》2021年第34期1-5,共5页Science and Technology Innovation Herald

摘　　要：蜂窝麻面是桥梁病害的重要表现形式之一,目前关于蜂窝麻面的自动识别检测技术比较有限。本文采用图像处理的方法对其进行分析,需要先对图像大小进行调整,然后根据预先设定的尺寸将图像划分成多个小图像块,将每个图像块转换为灰度图像,从中选取一些图像块作为样本进行训练学习,采用灰度共生矩阵的方法提取每个小图像块的纹理结构特征,计算有缺陷样本与无缺陷样本间的平均欧氏距离,然后根据欧氏距离阈值对小图像块是否有蜂窝麻面等缺陷做出判断,并计算缺陷面积。结果表明,该方法对缺陷的位置识别及面积计算都比较准确。Flaw is one of the important manifestations of bridge disease.At present,the automatic detection technology of flaw is limited and it is analyzed by image processing method.First,it may be necessary to adjust the image size and then divide it into a plurality of small image blocks according to a preset size.Each image block is transformed into a grayscale image and some image blocks are selected as training samples.The texture structure of each small image block is extracted by gray level co-occurrence matrix method.Calculating the average euclidean distance between the defective samples and the defect-free samples and judging whether the image block has flaw according to it and then calculating the flaw area.The results show that the method is accurate to identify the location and area of the flaw.

关 键 词：图像划分 灰度化 灰度共生矩阵 纹理结构特征 欧氏距离 位置识别 

分 类 号：U446[建筑科学—桥梁与隧道工程]