摘要
为快速、高密度地检测弹簧线径,提出一种基于机器视觉的压缩弹簧中段线径测量方法。介绍测量算法的整体结构,在获取弹簧边沿的关键拐点后将弹簧各段分开,保留近似直线的中段边沿部分并进行最小二乘直线拟合;根据拟合直线和关键拐点生成检测基线,按预设的检测密度生成检测线段,测得各检测点位的线径。在精度测量实验中,最大相对误差为2.84%。在不同检测密度下的运行速度对比实验中,算法平均运行时间为120.4~130.3 ms,检测密度相差16倍时检测速度相差7.60%。经过批量验证,该方法快速可靠,一次运行可在200 ms内完成弹簧中段位置数十至一千多个测量点的线径测量。
In order to detect wire diameter of spring quickly and densely,a middle section wire diameter measurement method of compression spring based on machine vision was proposed.The overall structure of the measurement algorithm was introduced.After obtaining key inflection points of the spring edge,each section of spring was separated,the edges of middle part closed to straight lines were reserved and the least square straight line fitting was carried out;the detection baselines were generated according to the fitted straight lines and the key inflection points,the detection line segments were generated according to preset detection density,and the wire diameter of each detection point was measured.In the precision measurement experiment,the maximum relative error was 2.84%.In the running speed comparison experiment under different detection densities,the average running time of the algorithm was 120.4~130.3 ms,when the difference of detection density was 16 times,the difference of detection speed was 7.60%.After batch verification,it is verified that the method is fast and reliable,and wire diameter measurement of dozens to more than one thousand measuring points in the middle of spring can be completed in one running within 200 ms.
作者
易焕银
刘志明
覃明俊
YI Huanyin;LIU Zhiming;QIN Mingjun(Department of Mechanical and Electrical Engineering,Guangdong Communication Polytechnic,Guangzhou Guangdong 510800,China)
出处
《机床与液压》
北大核心
2022年第21期125-131,共7页
Machine Tool & Hydraulics
基金
2020年度广东省普通高校青年创新人才类项目(2020KQNCX166)
2021年度广东大学生科技创新培育专项资金项目(pdjh2021b0779)
2020年度广东交通职业技术学院校级大学生科技创新项目(GDCP-ZX-2020-011-N2)。
关键词
机器视觉
压缩弹簧中段线径
高密度测量
Machine vision
Compression spring middle section wire diameter
High density measurement
