Image classification method and device and storage medium

Abstract

The invention relates to the technical field of intelligent decision making, and discloses an image classification method, which comprises the steps of obtaining training images, and preprocessing thetraining images to generate a first image set; training the convolutional neural network according to the first image set to obtain an image classification model, the image classification model comprising a first convolutional layer and a second convolutional layer, and the first convolutional layer and the second convolutional layer being convolutional layers formed by multiplying a residual module and an attention module; acquiring images to be classified, and preprocessing the images to be classified to generate a second image set; and classifying the images in the second image set according to the image classification model to obtain image categories, wherein the image categories include malignant images and benign images. According to the image classification method provided by the invention, the image classification model is optimized, and the accuracy of classifying the ultrasonic images by the image classification model is improved.

Description

technical field [0001] The invention relates to the technical field of intelligent decision-making, in particular to an image classification method, an electronic device and a computer-readable storage medium. Background technique [0002] The residual network (ResNet) in the convolutional neural network can build a deeper neural network by connecting features and improving shallow gradients. Although the neural network can be optimized through this high-dimensional nonlinear mapping method, it is difficult to achieve good results in the image classification of some small data image sets, especially in the classification of medical data image sets. In deep learning tasks, if the amount of data is too small, the residual network often cannot use too high-dimensional channels. At the same time, the number of layers of the neural network must be carefully controlled, because training a large neural network with a small amount of data will cause overfitting. leading to a decrea...

AI Technical Summary

Problems solved by technology

Although the neural network can be optimized through this high-dimensional nonlinear mapping method, it is difficult to achieve good results in the image classification of some small data image sets, especially in the classification of medical data image sets

In deep learning tasks, if the amount of data is too small, the residual network often cannot use too high-dimensional channels. At the same time, the number of layers of the neural network must be carefully controlled, because training a large neural network with a small amount of data will cause overfitting. leading to a decrease in image classification accuracy

Images