基于OpenCV的导向架表面缺陷检测系统

Surface Defect Detection System of Guide Frame Based on OpenCV

针对工厂人工目视检测导向架表面缺陷成本高、效率低等问题,提出采用机器视觉检测系统来提高检测效率和精度。通过自主搭建导向架表面缺陷自动检测平台,通过CMOS相机采集表面缺陷图像,通过Visual Studio 2017应用软件上运行的检测算法分析导向架表面缺陷。采用灰度处理、中值滤波、阈值选择、边缘检测和表面缺陷辨识等操作,对导向架表面缺陷进行识别分析并将结果反馈给执行机构,从而更加有效的识别次品。实验结果验证了该算法的可行性,单个工件的平均检测时间为43.24 ms,符合工厂的检测需求,由此可推断该检测平台及算法对导向架表面缺陷的高效检测具有一定的参考意义。

In view of the high cost and low efficiency of manual visual inspection of guide frame surface defects in factories,a machine vision inspection system is proposed to improve the inspection efficiency and accuracy.By building an automatic detection platform for the surface defects of the guide frame,the surface defect images are collected through a CMOS camera,and the surface defects of the guide frame are analyzed through the detection algorithm running on the Visual Studio 2017 application software.Using operations such as grayscale processing,median filtering,threshold selection,edge detection and surface defect identification,the surface defects of the guide frame are identified and analyzed,and the results are fed back to the actuator,so as to identify defective products more effectively.The experimental results verify the feasibility of the algorithm.The average detection time of a single workpiece is 43.24 ms,which meets the detection requirements of the factory.It can be concluded that the detection platform and algorithm have certain reference significance for the efficient detection of surface defects of the guide frame.