摘要
为应对自动化装配线质量检测高速与高精度的双重需求,基于机器视觉的智能检测系统创新融合多光源照明与多角度成像技术,集成深度学习算法,构建了产品质量的实时智能检测方案。在汽车零部件装配线实测中,该系统在200件/分钟的生产节拍下达到98.6%的检测准确率,漏检率稳定在0.1%以下。即使在装配线振动及环境光照波动等复杂工况下,系统仍保持稳定的高精度检测能力。
To meet the dual requirements of high speed and high precision in the quality inspection of automated assembly lines,an intelligent inspection system based on machine vision has innovatively integrated multi-light source illumination and multi-angle imaging technologies.By integrating deep learning algorithms,a real-time intelligent detection solution for product quality has been developed.In the actual measurement of automotive parts assembly lines,under a production cycle of 200 pieces per minute,this system achieves a detection accuracy of 98.6%,and the missed detection rate is stably maintained below 0.1%.Even under complex working conditions,such as vibrations of the assembly line and fluctuations in ambient illumination,the system still maintains a stable and high-precision detection ability.
作者
方赓
FANG Geng(Nanjing Polytechnic Institute,Nanjing Jiangsu 210000,China)
出处
《信息与电脑》
2025年第16期89-92,共4页
Information & Computer
关键词
机器视觉
自动化装配
质量检测
多视角成像
深度学习
machine vision
automated assembly
quality inspection
multi-view imaging
deep learning
