Research progress of bearing defect detection based on machine vision
Machine vision is a technology that uses machines to replace human eyes for measurement and inspection.This technology has the advantages of high efficiency,fast speed,and low cost when used in defect detection.Many scholars applied it in different fields(agriculture,aerospace,etc.)for defect detection and got better results.At present,it was also gradually adopted in bearing defect detection.Therefore,the bearing defect detection algorithms applied in different bearing defects,machine learning,and deep learning were reviewed,and the performance of defect detection algorithms was analyzed,summarized and compared.Firstly,the wear mechanism caused by bearing defects was discussed and analyzed,and the common wear forms of bearings(corrosion wear,fatigue wear,adhesive wear,raceway wear,etc.)were introduced in detail.Secondly,the differences and characteristics of detection algorithms based on machine learning and deep learning were respectively introduced.Then,the research,application and analysis of machine learning algorithms and deep learning algorithms for bearing defect detection were listed,which mainly included artificial neural networks(ANN),principal component analysis(PCA),support vector machines(SVM),etc.of machine learning and the application of single stage and two stage target detection algorithms of deep learning.Finally,in order to promote the use of deep learning algorithms for the diagnosis of bearing defects,the challenges and future research directions of bearing defect detection were presented and detailed suggestions were given for specific problems,and the current research status of machine vision in bearing defect detection was summarized and outlooked.
NETL
NSTL
万方数据
维普
