Defective gear classification method and device and computer readable storage medium

The invention relates to the technical field of gear surface defect detection, in particular to a defective gear classification method and device and a computer readable storage medium, and the method comprises the steps: constructing a defective gear depth migration network model which comprises a feature extractor, a domain condition channel attention module and a classifier; inputting a training set sample into a feature extractor, then inputting the training set sample into a domain condition channel attention module, and calculating distance loss of classification input features of a target domain and a source domain on a feature layer and distance loss of classification output of the target domain and the source domain on a classification layer; constructing a loss function, and updating network parameters of the model; and inputting the target domain sample without the label into the trained defect gear depth migration network model to obtain a classification result. According to the method, the target