基于机器视觉的轴承生产中的表面质量在线检测与研究

摘要

数字图像处理技术与模式识别技术结合，可广泛应用于工业产品的分类识别。针对轴承在线装配生产中存在的缺陷检测识别的需求，研究基于机器视觉的自动检测系统，采用面阵相机对不同表面进行对比测量，对采集的图像进行二值化处理，为后续的图像预处理、模式识别、特征提取及特征值对比做了分析和准备，以实现产品的在线表面缺陷检测识别。%The combining of digital image processing technique and pattern recognition technique can be widely used in industry products classification and recognition. For the defects detection and identification needs in bearing assembly line,automatic detection system based on machine vision was studied,contrast measurement on different surfaces was made by plane array camera. Binaryzation processing was done for acquired image. It provides foundation for subsequent image pretreatment,pattern recognition,feature extrac-tion and eigenvalue comparison,to realize product line surface defect detection and identification.