摘要
针对传统线缆绝缘厚度测量方法存在效率低和准确度差的问题,提出一种基于自适应局部交替遗传算法(ALA-GA)的绝缘厚度检测方法;该方法利用ALA-GA算法在试件图像的内外边缘交替搜索从而获得最优绝缘厚度位置;该算法引入试件先验结构知识,根据试件截面边缘曲率特征自适应选取初始种群,从而保证初始种群基因的优质性和多样性;将交叉和变异操作置前,对于试件截面内外边缘局部交替地自适应改变交叉和变异的方式,从而提高遗传算法的求解速度;为了不丢失任一边缘的优质基因,对交叉、变异后得到的新种群和原种群共同执行后置选择操作;每获得一个最优检测位置,剔除该位置附近的其余解,如此迭代执行ALA-GA算法以获得精确的绝缘厚度检测结果;对比实验以及能力验证表明,基于ALA-GA方法的时间代价为0.6~0.7 s,最薄点测量误差为0.001 2~0.001 5 mm,平均测量误差为0.001 3~0.001 7 mm,测量重复性为0.001 8~0.002 1 mm,均优于现有先进方法,且对不规整线缆泛化能力良好。
Aiming at the low efficiency and poor accuracy of traditional cable insulation thickness measurement methods,an insulation thickness detection method based on adaptive local alternating genetic algorithm(ALA-GA) is proposed.The method uses the ALA-GA algorithm to alternately search the inner and outer edges of the specimen image so as to obtain the optimal insulation thickness position;The algorithm introduces the priori structural knowledge of the specimen,adaptively selects the initial population according to the curvature characteristics of the specimen cross-section edges,and ensures the high quality and diversity of the initial population genes.The crossover and mutation operations are placed in the front,and the crossover and mutation modes are changed locally and adaptively for the inner and outer edges of the specimen section,so as to improve the solution speed of the genetic algorithm;In order not to lose the high-quality genes of any edge,the post-selection operation is achieved by the original population and new population obtained after the crossover and mutation;An optimal detection position is obtained every time,and other solutions near the position are eliminated,the ALA-GA algorithm is iterated to obtain the result of accurate insulation thickness detection.Comparison experiments and capability verification show that the ALA-GA-based method has a time cost of 0.6~0.7 s,a thinnest point measurement error of 0.001 2~0.001 5 mm,an average measurement error of 0.001 3~0.001 7 mm,and a measurement repeatability of 0.001 8~0.002 1 mm,the method is superior to existing advanced methods,and has a good general capability for irregular cables.
作者
刘付渝杰
马宏园
许浩然
宋俊儒
罗睿
李杨
LIUFU Yujie;MA Hongyuan;XU Haoran;SONG Junru;LUO Rui;LI Yang(School of Computer Science and Technology,Guangdong University of Technology,Guangzhou 510006,China;Quality&Metrology Supervision Testing Institute,Maoming 525000,China;School of Intelligent Systems Engineering,Sun Yat-Sen University,Shenzhen 518107,China;Pengcheng Laboratory,Shenzhen 518055,China;Maoming Branch of Guangdong Institute of Special Equipment Inspection and Research,Maoming 525000,China;Maoming Material Reserve Center,Maoming 525000,China)
出处
《计算机测量与控制》
2024年第8期55-63,71,共10页
Computer Measurement &Control
基金
广东省市场监督管理局科技项目(2024CZ11)
广东省茂名市科技计划项目(220420094550422)。
关键词
机器视觉
遗传算法
光缆电缆
自动测量
绝缘厚度
检验检测
machine vision
genetic algorithm
electric or optical cable
automatic measurement
insulation thickness
inspection and testing