The best way to conference proceedings by Francis Academic Press

Web of Proceedings - Francis Academic Press
Web of Proceedings - Francis Academic Press

Research on the Application of Deep Learning Algorithms on Image Classification

Download as PDF

DOI: 10.25236/mmmce.2021.036

Author(s)

Zijian Wang

Corresponding Author

Zijian Wang

Abstract

In recent years, the research of deep learning (DL) has received widespread attention. Compared with the traditional shallow network model, the multi-layer network structure of the deep learning model can express complex functions more effectively, thereby learning features with stronger characterization ability, and applying deep learning to image classification can effectively improve Classification accuracy. However, deep learning still faces severe challenges in practical applications: the important basic unit model of deep learning-Restricted Boltzmann Machine (RBM) has high computational complexity and model in the training process. The likelihood of training data is not high enough, which leads to a long training time for deep learning models.

Keywords

Deep learning algorithms, Image classification, Application study