Abstract: A document classification method includes: obtaining a document; extracting document keywords and a document abstract from the document; determining, according to the document keywords and the document abstract, a first classification label of the document; receiving a second classification label of the document, the second classification label being determined according to the document keywords, the document abstract, and the first classification label; obtaining a classification efficiency value of the documents, the classification efficiency value indicating a confidence level of the second classification label; and determining the second classification label as a final classification label of the document if the classification efficiency value is greater than or equal to a first threshold. The document classification method described above may improve the speed of document classification by computer, thereby improving the efficiency of document classification.

Abstract: Disclosed is a method for recognizing distribution network equipment based on Raspberry Pi multi-scale feature fusion. The method includes obtaining an initial sample data set; constructing an object detection network composed of EfficientNe-B0 backbone network, multi-scale feature fusion module and a regression classification prediction head; training the object detection network by taking the initial sample data set as a training sample; finally, detecting inspection pictures by using a the trained object detection network. A light-weight EfficientNet-B0 backbone network feature extraction method obtains more features of objects. Meanwhile, an introduction of multi-scale feature fusion better adapts to small object detection, and a light-weight y_pred regression classification detection head is effectively deployed and realized in Raspberry Pi embedded equipment with tight resources and limited computing power.