Method and system for rapidly detecting welding spot defects of SMT production line based on improved Yolox

The invention discloses an SMT production line welding spot defect rapid detection method and system based on improved Yolox, and the method comprises the steps: collecting welding spot defect image data stored by production line AOI equipment, carrying out the preprocessing, and dividing a training set and a test set according to a proportion; a Yolox neural network structure is improved, size output by a backbone part after model improvement is adjusted so as to be matched with a check network part, and an improved Yolox neural network model is obtained; performing model training and testing on the improved Yolox neural network by using the training set and the testing set to obtain an SMT production line welding spot defect target detection model; wherein the input of the SMT production line welding spot defect target detection model is a PCBA image, and the output is defect positioning and classification; and the detection model is embedded into the AOI system of the SMT production line, so that the deplo