曾理等-工业CT图像中小间隙裂纹亚像素测量方法.docVIP

工业CT图像中小间隙裂纹的亚像素测量方法 曾理①③, 安贝贝①②, 悦秀娟①③ ① 重庆大学 光电技术及系统教育部重点实验室 ICT研究中心 重庆 400044； ②重庆市环境科学研究院? 重庆 400020 ③重庆大学 数理学院 重庆 400044 drlizeng@ ( 摘要：工业部件内部小间隙裂纹测量是公认的技术难题，而基于工业CT断层图像的裂纹测量的基础是裂纹边缘检测。脊波变换是近年在小波变换基础上发展起来的新的分析方法，脊波变换对于图像中的线状奇异性（如边缘、裂纹等）较敏感，且它对图像中的点状噪声有抑制作用。本文在利用脊波变换检测工业CT图像中工件裂纹边缘的基础上，对小间隙裂纹的平均宽度进行亚像素测量。而工业CT图像成像时存在的点扩散效应，使测得裂纹宽度较真实的要宽些。因此本文采用宽度收缩的方法获得裂纹平均宽度并进行尺寸标定。实验结果表明，基于脊波变换的方法能够使小间隙裂纹宽度测量的精度达到亚像素级。 关键词：图像处理；脊波变换；亚像素；裂纹测量；工业CT 中图分类号：TP391 文献标识码：A The sub-pixel measurement method for narrow cracks in ICT image ZENG Li①③ ，AN Bei-bei①② ， YUE Xiu-juan①③ ① ICT Research Center, Key Laboratory of Optoelectronic Technology and System of the Education Ministry of China, Chongqing University, Chongqing 400044, China; ② Chongqing Institute of Environmental Science, Chongqing, 400020, China; ③College of Mathematics and Physics, Chongqing University, Chongqing 400044, China Abstract: This paper is mainly to solve the problem of measuring the small crack in industrial CT images. In the process of structural materials working, cracks often appear, the existence of cracks greatly influences the material’s toughness, strength, hardness, and other mechanical properties. The crack in the workpiece is divided into two parts, one is internal crack, the other is external crack, the external crack is very intuitionistic, this paper mainly research the small internal cracks. Because the crack is very small, less than 1 pixel, and the influence of noise, the measurement is very difficult. Computer tomography (CT) is an important non-destructive testing technology, and it can describe the detail information of workpiece’s internal structure. The CT detection cannot be influenced by the material type, shape structural, its imaging is visual and its resolving power is good, especially in the testing of intricate workpiece. But there is need to measure the crack by rule and line. In this paper, we use Ridgelet transform to detect the edge. Ridgelet transform was developed from wav