Research on Image Classification Based on Convolutional Neural Network

Abstract

In the field of image research, aiming at the problems of complexity, large amount of calculation and low accuracy in the traditional image classification process, a variety of machine learning algorithms can be used. By extracting image features, the computer can effectively manage and classify different types of images. In recent years, convolutional neural networks have gradually become the mainstream of image classification applications, and performed very well in the field of image classification. Based on the TensorFlow deep learning framework, a 9-layer convolutional neural network was designed in this study, we applied the Modified National Institute of Standards and Technology (MNIST) image dataset to train the network model and optimize model parameters, and compared the classification effect with the Support Vector Machine (SVM) model. The results show that the classification accuracy of convolutional neural network is 4% higher than that of SVM model.