Deep learning method for detecting surface defects of filter screen

The invention provides a deep learning method for filter screen surface defect detection, which comprises the following steps that an attention mechanism SE module is added to the last layer of a backbone network of a YOLOv5 network algorithm, and the attention mechanism SE module is used for adaptively adjusting and correcting feature weights in a feature extraction process; a feature pyramid network BiFPN module is introduced into a neck network of the YOLOv5 network algorithm, an original path aggregation network PANet structure of the YOLOv5 network algorithm is removed, and an optimized YOLOv5 network algorithm is obtained; and training the optimized YOLOv5 network algorithm through the filter screen surface defect data set, and obtaining a YOLOv5 target detection network for filter screen surface defect detection. An attention mechanism SE and a pyramid network BiFPN module are added in a YOLOv5 network algorithm, so that the defects on the surface of the filter screen can be accurately detected. 本发明提供一