摘要
针对传统织针加工行业织针指标检测统一性不高的问题,设计一种基于图像处理的圆机织针关键尺寸参数测量系统,可实现针身多部位尺寸的在线测量。基于机器视觉与图像处理技术,使用高分辨视觉系统获取织针原始图像,预处理后从其图像像素点入手,寻找关键尺寸边界像素跳变点,最后计算得到织针针身、针钩、动程及针踵等目标尺寸。利用该系统重复测量多组人工优选的统一性较好的织针,并与人工检测进行对比。结果表明:系统准确稳定,单枚织针检测耗时约45 ms,优于人工;系统测量与人工检测结果对比,织针关键尺寸测量结果平均误差为0.0468 mm,对应平均像素点误差约为1.0000个像素点。
In view of the lack of detection methods in the traditional knitting needle processing industry and the current situation of poor uniformity in the detection of knitting needle technical indicators,a measurement system for the key dimensions of circular knitting needles based on image processing technology is designed,which can realize online measurement of the dimensions of multiple parts of the needle body.Based on machine vision and image processing technology,a high-resolution vision system is used to obtain the original image of the knitting needle.After preprocessing,it starts from its image pixels to find the key dimension boundary pixel change points.Finally,the target dimension of the needle body,hook,stroke and butt of the knitting needle is calculated and converted,and it can judge whether it meets the technical index requirements.The designed dimension measurement system is used to repeatedly measure multiple groups of manually selected knitting needles with better uniformity,and then compares with the manual detection results to verify the accuracy and reproducibility of the system.Experimental results show that the system is accurate and stable and meets the requirements.The detection and measurement of a single needle takes about 45 ms,which is faster than manual.The comparison between the system measurement result and the manual detection result shows that the average error of the measurement result of the key dimension of the knitting needle is 0.0468 mm,and the corresponding average pixel error is about 1.0000 pixel.
作者
库朋涛
张周强
胥光申
刘学靖
Ku Pengtao;Zhang Zhouqiang;Xu Guangshen;Liu Xuejing(School of Mechanical and Electrical Engineering,Xi’an Polytechnic University,Xi’an,Shaanxi 710600,China;Xi’an Key Laboratory of Modern Intelligent Textile Equipment,Xi’an,Shaanxi 710048,China)
出处
《针织工业》
北大核心
2022年第3期1-5,共5页
Knitting Industries
基金
国家自然科学基金青年项目(61701384)
陕西省自然科学基金(2020JQ-820)
陕西省教育厅重点科学研究计划项目(20JS051)
西安现代智能纺织设备重点实验室(2019220614SYS021CG043)
西安工程大学柯桥纺织产业创新研究院项目(19KQYB03)
关键词
图像处理
机器视觉
圆机
织针
尺寸测量
像素点
Image Processing
Machine Vision
Circular Machine
Knitting Needle
Dimension Measurement
Pixel
作者简介
库朋涛(1994—),男,硕士研究生。主要从事结构设计与机器视觉研究;通讯作者:张周强(1983—),男,副教授,博士。E-mail:Zhangzhouqiang208@126.com。
