Exploring quality inspection and grade judgment of distortion defects in the fabrication of spectacle lenses

IF 1.4 Q3 OPERATIONS RESEARCH & MANAGEMENT SCIENCE
Hong-Dar Lin, Tung-Hsin Lee, Chou-Hsien Lin, Y. Chiu
{"title":"Exploring quality inspection and grade judgment of distortion defects in the fabrication of spectacle lenses","authors":"Hong-Dar Lin, Tung-Hsin Lee, Chou-Hsien Lin, Y. Chiu","doi":"10.5267/j.dsl.2022.6.001","DOIUrl":null,"url":null,"abstract":"This study explores the quality control system featuring visual inspection and grade judgment for detecting distortion defects in spectacle lens fabrication. Spectacle lenses must be precisely curved to help accommodate nearsightedness and farsightedness. The curved shape allows the lens to have different curvatures in different areas during grinding. The spectacle lens will be prone to optical distortion when the curvature changes abnormally. Accordingly, this study proposes an automatic distortion flaw inspection system for spectacle lenses to substitute professional inspectors who rely on empirical judgment. We first apply the digital imaging of a concentric circle pattern through a testing lens to create an image of that lens. Second, the centroid–radii model is employed to stand for the shape property of each concentric circle in the image. Third, by finding the deviations of the centroid radii for detecting distortion flaws through a small variation control method, we obtain a different image showing the detected distortion regions. Four, based on the distortion amounts and locations, we establish the fuzzy membership functions and inference rulesets to measure distortion severity. Finally, the GA-ANFIS model is applied to determine the quality levels of distortion severity for the detected distortion flaws. Trial outcomes reveal that the proposed automatic inspection system can help practitioners in spectacle lens fabrication, for it attains a high 94% correct classification rate of quality grades in distortion severity, 81.09% distortion flaw detection rate, and 10.94% fake alert rate, in distortion inspection of spectacle lenses.","PeriodicalId":38141,"journal":{"name":"Decision Science Letters","volume":"33 1","pages":""},"PeriodicalIF":1.4000,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Decision Science Letters","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.5267/j.dsl.2022.6.001","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"OPERATIONS RESEARCH & MANAGEMENT SCIENCE","Score":null,"Total":0}
引用次数: 1

Abstract

This study explores the quality control system featuring visual inspection and grade judgment for detecting distortion defects in spectacle lens fabrication. Spectacle lenses must be precisely curved to help accommodate nearsightedness and farsightedness. The curved shape allows the lens to have different curvatures in different areas during grinding. The spectacle lens will be prone to optical distortion when the curvature changes abnormally. Accordingly, this study proposes an automatic distortion flaw inspection system for spectacle lenses to substitute professional inspectors who rely on empirical judgment. We first apply the digital imaging of a concentric circle pattern through a testing lens to create an image of that lens. Second, the centroid–radii model is employed to stand for the shape property of each concentric circle in the image. Third, by finding the deviations of the centroid radii for detecting distortion flaws through a small variation control method, we obtain a different image showing the detected distortion regions. Four, based on the distortion amounts and locations, we establish the fuzzy membership functions and inference rulesets to measure distortion severity. Finally, the GA-ANFIS model is applied to determine the quality levels of distortion severity for the detected distortion flaws. Trial outcomes reveal that the proposed automatic inspection system can help practitioners in spectacle lens fabrication, for it attains a high 94% correct classification rate of quality grades in distortion severity, 81.09% distortion flaw detection rate, and 10.94% fake alert rate, in distortion inspection of spectacle lenses.
探索镜片制造中变形缺陷的质量检测与等级判断
本研究探索以视觉检测与等级判断为核心的质量控制体系,用于检测镜片制造过程中的畸变缺陷。眼镜镜片必须精确弯曲,以帮助适应近视眼和远视眼。曲面形状使得透镜在磨削过程中,在不同的区域具有不同的曲率。当曲率发生异常变化时,镜片容易发生光学畸变。据此,本研究提出了一种眼镜镜片畸变缺陷自动检测系统,以替代依赖经验判断的专业检测人员。我们首先通过测试透镜应用同心圆图案的数字成像来创建该透镜的图像。其次,采用质心-半径模型表示图像中各同心圆的形状属性;第三,通过小变化控制方法找到检测畸变缺陷的质心半径的偏差,得到显示检测畸变区域的不同图像。第四,根据变形量和位置,建立模糊隶属函数和推理规则集来度量变形的严重程度。最后,应用GA-ANFIS模型对检测到的畸变缺陷确定畸变严重程度的质量等级。试验结果表明,所提出的自动检测系统对眼镜镜片的变形严重程度质量等级的正确分类率高达94%,变形探伤率高达81.09%,假报警率高达10.94%,可以为眼镜镜片制造从业者提供帮助。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Decision Science Letters
Decision Science Letters Decision Sciences-Decision Sciences (all)
CiteScore
3.40
自引率
5.30%
发文量
49
审稿时长
20 weeks
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信