基于灰度共生矩阵的表面粗糙度研究  

Implementation of Surface Roughness With Image Texture Analysis

作　　者：冯建[1] 周晨波[1] 于文英[1] 许泓彧[1] 董燕[1] 

机构地区：[1]烟台大学光电信息科学技术学院,烟台264005

出　　处：《光学与光电技术》2007年第2期39-41,共3页Optics & Optoelectronic Technology

摘　　要：介绍了激光散斑的形成及其统计性质,拍摄大量不同加工工艺、不同表面粗糙度标准样块的散斑图像,基于激光散斑图像的灰度共生矩阵及惯性矩、角二阶矩、熵、相关性4个参量,对图像进行了纹理分析。分别绘制了不同加工工艺、不同表面粗糙度标准样块的4个参量与表面粗糙度关系的特性曲线。最后将实验结果和未知表面粗糙度工件的实验数据进行比对,准确估算出了工件的表面粗糙度。The formation and statistical characteristic of laser speckle are introduced. The speckle images of samples with different processing and different surface roughness are taken. Based on the gray level co-occurrence matrix, moment of inertia, entropy, angle second moment of laser speckle, the texture of images is analyzed. The relation curves of the four parameters and surface roughness are drawn. The experimental results and the data of the object with unknown surface roughness are compared, and the surface roughness of the object are accurately estimated.

关 键 词：激光散斑 纹理 灰度共生矩阵 表面粗糙度 

分 类 号：O436.2[机械工程—光学工程]