Image classification method based on convolutional neural network

Abstract

The invention discloses an image classification method based on a convolutional neural network. The method comprises the following steps: A, collecting images through an image collector, and saving the image to a memory, and performing the encryption processing on the same; B, inputting the collected images to a pre-training model after adjusting the collected images into the uniform size; C, training the images to compute an average image in a training set, performing network parameter training after subtracting the average image from each image in the image training set, thereby obtaining aconvolutional neural network model; and D, performing a classification operation on the trained images, namely accomplishing the classification of the image. The image classification method provided by the invention is high in efficiency and capable of effectively improving the accuracy and efficiency of the image classification.

Description

technical field [0001] The invention relates to the technical field of image classification, in particular to an image classification method based on a convolutional neural network. Background technique [0002] In recent years, with the emergence of millions of labeled training sets and the emergence of GPU-based training algorithms, it is no longer a luxury to train complex convolutional network models. Convolutional neural networks are an efficient method that has gradually developed and attracted widespread attention. image recognition method. A large number of models based on convolutional neural networks have achieved good results in handwritten font recognition and classification tests of the ImageNet library. Many papers have used convolutional neural networks to achieve good results in visual classification tasks. [0003] The basic structure of the convolutional neural network includes two layers, one is the feature extraction layer, the input of each neuron is co...

AI Technical Summary

Problems solved by technology

Due to the huge parameters of the convolutional neural network, the number of network parameters ranges from several megabytes to hundreds of megabytes, resulting in long neural network training time, consuming a lot of memory space, and relatively high requirements for the number of training samples. In addition, the security is poor, and for Massive image data cannot be quickly classified and processed, and the efficiency is low

Images