Abstract:With the rapid development of modern manufacturing, traditional testing methods have been unable to meet the needs 〖JP2〗of the market. Machine vision plays an invaluable role in bearing inspection. It has the advantages of non-contact, accuracy, objectivity〖JP〗, repeatability, real-time and automation, and has broad application prospects and research value. On the basis of reading a large number of documents, starting from three aspects of size detection, surface defect detection and image processing, the research of predecessors is summarized, and the research status and development trend of machine vision are expounded comprehensively in bearing detection, which can provide reference for relevant research in the later stage.