Image classification method and device based on deep learning, server and medium

Abstract

The invention relates to an intelligent decision, and discloses a deep learning-based image classification method, and the method comprises the steps of constructing a first classification model, training the first classification model to obtain an adjusted first classification model, and inputting a first sample into the adjusted first classification model to obtain first feature data; compressing the structure of the first classification model to obtain a second classification model, sending the first feature data and the second classification model to each second server, receiving second parameters fed back by each second server, obtaining an updated first classification model based on the second parameters, and training the updated first classification model to obtain a target classification model; and inputting the to-be-classified image into the target classification model to obtain an image classification result. The invention further provides an image classification device, a server and a medium. According to the invention, the model training efficiency is improved, and the image classification accuracy is improved.

Description

technical field [0001] The present invention relates to the field of intelligent decision-making, in particular to an image classification method, device, server and medium based on deep learning. Background technique [0002] With the development of artificial intelligence, the application of deep learning is becoming more and more extensive. For example, a large bank unites several small banks to train a public classification model through deep learning to classify images. The local databases of each bank store user data in the form of images (images of certificates retained by users when handling business, images collected on-site) , business content image, receipt image, etc.). [0003] In deep learning, since the models corresponding to each small bank are relatively large, the computing power requirements for small banks are relatively high, and the potential computing power of large banks is not fully utilized, resulting in low model training efficiency and low train...

AI Technical Summary

Problems solved by technology

[0003] In deep learning, since the models corresponding to each small bank are relatively large, the computing power requirements for small banks are relatively high, and the potential computing power of large banks is not fully utilized, resulting in low model training efficiency and low training efficiency. The obtained model parameters are inaccurate, resulting in low image classification accuracy of the public classification model

Images