现代制造工程Issue(4)：123-131,9.DOI:10.16731/j.cnki.1671-3133.2025.04.015
基于机器视觉的套筒件尺寸测量研究
Research on dimensional measurement of sleeve pieces based on machine vision
摘要
Abstract
Aiming at the problems of low efficiency and low precision of measuring sleeve size in traditional industrial inspection,a sleeve size measurement method based on machine vision was proposed.Firstly,the sleeve image was captured by the CMOS camera calibrated by Halcon software,and pre-processed by bilateral filtering and threshold segmentation.The improved Canny algorithm combined with the optimal segmentation threshold obtained by the Otsu method was used to obtain the pixel-level coor-dinates of the edges.The improved Zernike moments algorithm was used to locate the edges accurately.Finally,the inner and out-er diameters of the sleeve were solved by fitting the inner and outer circles using the least squares method.The experimental re-sults show that the average measurement accuracy of the method is less than 0.02 mm,and the measurement accuracy and effi-ciency meet the requirements of enterprises for rapid and accurate inspection.关键词
机器视觉/套筒/亚像素/尺寸测量Key words
machine vision/sleeve/sub-pixel/dimensional measurement分类
信息技术与安全科学引用本文复制引用
蔡舒,吴超华,罗威,史晓亮..基于机器视觉的套筒件尺寸测量研究[J].现代制造工程,2025,(4):123-131,9.基金项目
国家自然科学基金项目(52375201) （52375201）
