基于深度学习的机器视觉轴承检测方法研究  

Research on Bearing Detection Method of Machine Vision Based on Deep Learning

作　　者：柳清星 姚虹光 秦超 谢广玲 陈雨虹 LIU Qingxing;YAO Hongguang;QIN Chao;XIE Guanging;CHEN Yuhong(Guangxi University of Finance and Economics,Nanning Guangxi 530003,China)

机构地区：[1]广西财经学院,广西南宁530003

出　　处：《信息与电脑》2023年第6期50-52,共3页Information & Computer

摘　　要：轴承缺陷检测是机器视觉技术中一个重要的应用领域,传统算法需针对特征设计特殊算子检测缺陷,算法比较复杂,局部算子实现困难,大大降低了算法的稳定性,开发效率不高。基于此,首先分析利用机器视觉检测缺陷的方法,其次分析深度学习在轴承缺陷检测中的应用,最后分析深度学习在轴承检测应用中的未来研究方向。Bearing defect detection is an important application field in machine vision technology.Traditional algorithms require special operators to detect defects based on features,resulting in high algorithm complexity and difficulty in implementing local operators,greatly reducing algorithm stability and low development efficiency.Based on this,first analyze the methods of using machine vision to detect defects,then analyze the application of deep learning in bearing defect detection,and finally analyze the future research directions of deep learning in bearing detection applications.

关 键 词：缺陷检测 机器视觉 深度学习 轴承 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]