李小菁
（广东省计量科学研究院，广东 广州 510405）
摘要：针对目前大多数小型零件尺寸测量采用人工方式存在效率低、精度差的问题，设计一套基于机器视觉的小型零件尺寸测量系统。该测量系统的硬件部分由工业相机、远心镜头和背光光源等组成；测量软件实现小型零件图像平滑滤波、自适应阈值二值化、感兴趣区域自动提取、亚像素边缘特征提取和最小二乘法曲线拟合等功能。通过对短U弯制小型零件进行测量实验，验证了该测量系统可实现小型零件多个几何量参数的一键快速测量，最大误差为0.021 mm，测量精度可达0.02 mm。
关键词：机器视觉；图像处理；尺寸测量；小型零件
中图分类号：TP271 文献标志码：A 文章编号：1674-2605(2023)02-0003-05
DOI：10.3969/j.issn.1674-2605.2023.02.003
Small Part Size Measurement System Based on Machine Vision
LI Xiaojing
(Guangdong Provincial Institute of Metrology, Guangzhou 510405, China)
Abstract: A machine vision based small part size measurement system is designed to address the issues of low efficiency and poor accuracy in manual measurement of most small parts. The hardware part of the measurement system consists of an industrial camera, a telecentric lens, and a backlight light source; The measurement software achieves functions such as smooth filtering of small part images, adaptive threshold binarization, automatic extraction of regions of interest, sub pixel edge feature extraction, and least squares curve fitting. By conducting measurement experiments on short U-shaped small parts, it was verified that the measurement system can achieve quick measurement of multiple geometric parameters of small parts with one click, with a maximum error of 0.021 mm and a measurement accuracy of 0.02 mm.
Keywords: machine vision; image processing; size measurement; small parts
