基于机器视觉的平板背板辅料检测系统

Machine Vision Based Inspection System for Accessories of Flat Back Panel

检测辅料是否贴装正确,是平板背板生产过程中的一道重要的工序。针对传统的人工肉眼检测平板背板辅料存在的效率低、准确率低、不能长时间作业等问题,开发了一套基于机器视觉技术的平板背板辅料自动检测系统,主要采用工业相机、镜头、工业光源、机械传动装置、图像采集卡、工控机和上位机软件等构成。研究设计了系统整体的机械结构方案;对系统的硬件进行选型,然后根据图像特点采用灰度变换和中值滤波算法增强对比度以及滤除噪声,ROI提取与阈值分割、连通域分析等算法进行检测;最后对系统检测软件进行设计。与传统的人工检测相比,采用本视觉系统进行检测能大大缩短检测时间,并能大幅度提高检测准确度。实验数据表明:开发的系统对每一个平板背板的检测时间大约为0.5 s,检测准确率能达到99.8%。

It is an important process in the production of flat panel backplane to detect whether the auxiliary materials are correctly pasted.Aiming at the problems of low efficiency, low accuracy, and unable to work for a long time in the traditional manual visual inspection of the auxiliary materials of the flat back plate, an automatic inspection system for the auxiliary materials of the flat back plate based on machine vision technology was developed, which was mainly composed of industrial camera, lens, industrial light source, mechanical transmission device, image acquisition card, industrial computer and upper computer software. The mechanical structure scheme of the whole system was studied and designed. Secondly, the hardware of the system was selected. According to the characteristics of the image, gray transformation and median filtering algorithm were used to enhance contrast and filter noise, ROI extraction and threshold segmentation, connected component analysis and other algorithms were used for detection. Finally, the system detection software was designed. Compared with the traditional manual detection, the visual system can greatly shorten the detection time and improve the detection accuracy. The experimental data shows that the detection time of this system for each flat backplane is about 0.5 s, and the detection accuracy can reach 99.8%.