The Algorithm Research of Image Classification Based on Deep Convolutional Network

Abstract

In order to solve large amount of images classification issues, a method is introduced by combining with Convolutional Neural Network(CNN) and IMVCl-10. Based on a large number of training data, the classification accuracy of the Convolution Neural Network can exceed the SVM classification accuracy. In this dataset, 3K and 100K batch data are used to train the depth convolution network, and the classification accuracy of 70% and more than 80% can be achieved.