基于高光特征的表面粗糙程度检测  

Surface Roughness Detection Based on Highlight Feature

作　　者：张柏行 李明心 林靖宇[1] ZHANG Baihang;LI Mingxin;LIN Jingyu(School of Electrical Engineering,Guangxi University,Nanning 530004)

机构地区：[1]广西大学电气工程学院,南宁530004

出　　处：《计算机与数字工程》2023年第8期1826-1831,共6页Computer & Digital Engineering

基　　金：国家自然科学基金项目(编号:61561005);国家留学基金项目(编号:201906660002);广西研究生教育创新计划项目(编号:YCSW2019026)资助。

摘　　要：通过分析高光特征与表面粗糙程度的关系,提出了一种基于高光特征的表面粗糙程度检测方法。高光存在会使图像局部产生颜色畸变导致信息丢失,通常被当作噪声筛除。但高光区别于过曝,也能通过亮度震荡的形式反映表面粗糙程度。针对这一信息,提出高光特征的概念;对比经典的光照模型,搭建针对高光特征这一局部特征的Highlight模型。通过最小二乘法多项式曲线拟合得到亮度趋势拟合曲线,Highlight模型检测高光特征,使其应用于表面粗糙程度检测。对瓷砖的检测实验表明了基于高光特征的表面粗糙程度检测的有效性。By analyzing the relationship between the characteristics of high light and the surface roughness,a new method of surface roughness detection based on the feature of high light is proposed.The existence of highlight will cause local color distortion of the image,resulting in the loss of information,which is usually regarded as noise screening.However,different from overexposure,highlight can also reflect the surface roughness in the form of brightness oscillation.In view of this information,the concept of highlight feature is proposed.Compared with the classical lighting model,the highlight model for the local feature of highlight feature is built.The brightness trend fitting curve is obtained by polynomial curve fitting with least square method,and the highlight model is used to detect the highlight feature,so that it can be applied to the surface roughness detection.The experiment of tile detection shows the effectiveness of surface roughness detection based on highlight feature.

关 键 词：高光特征 Highlight模型 光照模型 表面粗糙程度 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]