Study on image classification with convolution neural networks

Abstract

Classification is one of the most popular topics in image process. In this paper, it provides the specific process of convolutional neural network in deep learning. Here builds typical convolution neural network which parameters and the connection mode can be adjusted. In addition, we present our preliminary classification results. Through the experiments of a convolutional neural network on the Mixed National Institute of Standards and Technology database (MNIST), we compared with the classification results and analyzed in the experimental results with the parameters. The experimental results show that image classification effect is very good used by convolutional neural network.