Research on Accuracy Improvement Technology of Defect Detection Based on Machine Learning

Automated Optical Inspection (AOI) technology is widely used in industrial scenes, and the research mostly focuses on theoretical models or system design. In the industrial production environment, for different types and specifications of products, it is necessary to adjust the threshold of the detection parameters in time according to the detection of product defects, in order to improve the detection rate of the equipment. In this paper, in the study of the actual scene, the defect data generated by equipment detection is collected and collated, and the detection parameters of different types of defects are analyzed by machine learning methods, so as to improve the accuracy of defect detection.