机织钩面粘扣带表面质量的机器视觉检测方法  

Machine vision inspection method for surface quality of woven hook-and-loop fasteners

作　　者：金守峰[1] 李政 徐畅 昝杰 郭彩霞 JIN Shoufeng;LI Zheng;XU Chang;ZAN Jie;GUO Caixia(College of Mechanical and Electrical Engineering,Xi′an Polytechnic University,Xi′an,Shaanxi 710600,China;Hubei Key Laboratory of Digital Textile Equipment,Wuhan Textile University,Wuhan,Hubei 430200,China;Shaanxi Key Laboratory of Advanced Manufacturing and Evaluation of Robot Key Components,Baoji University of Arts and Sciences,Baoji,Shaanxi 721016,China)

机构地区：[1]西安工程大学机电工程学院,陕西西安710600 [2]武汉纺织大学湖北省数字化纺织装备重点实验室,湖北武汉430200 [3]宝鸡文理学院陕西省机器人关键零部件先进制造与评估省市共建重点实验室,陕西宝鸡721016

出　　处：《毛纺科技》2024年第4期103-110,共8页Wool Textile Journal

基　　金：陕西省自然科学基础研究计划项目(2023-JC-YB-288);湖北省数字化纺织装备重点实验室开放课题(KDTL2020005)。

摘　　要：针对机织钩面粘扣带表面质量人工检测中存在的精度低等问题,提出了基于机器视觉的机织钩面粘扣带表面质量检测方法。通过单目视觉的切向成像获得粘扣带表面的单行图像,避免了粘扣带相邻行之间的干扰。对单行粘扣带图像进行预处理后,采用投影法对其进行水平投影和垂直投影,提取粘扣带表面单行图像中的每个钩扣状区域,基于连通域统计法判别钩扣状区域的切割状态,计算其切开率,通过统计数据表征粘扣带表面质量。实验结果表明,该方法对机织钩面粘扣带表面质量的检测相对误差为1%,具有较高的稳定性。In response to the problems of low accuracy and low efficiency in manual inspection of the surface quality of woven hook-and-loop fasteners,a machine vision-based inspection method for the surface quality of woven hook-and-loop fasteners was proposed.The single-row image of the surface of the hook-and-loop fastener was obtained by tangential imaging of monocular vision,which avoids the interference between adjacent rows of the fastener.After pre-processing the single-row snap tape image,the horizontal projection and vertical projection were used to extract each hook-and-loop area in the single-row image of the snap tape surface,discriminate the cutting state of the hook-and-loop area based on the connected-domain statistical method,calculate its cutting rate,and characterize the surface quality of the snap tape by statistical data.The experimental results show that the relative error of the method on the surface quality of woven hook-and-loop fasteners is 1%,which has higher stability.

关 键 词：机织钩面粘扣带 切向成像 投影法 钩扣状区域 连通域 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]