Research on Image Classification Based on ResNet

DOI

10.2991/978-94-6463-300-9_98How to use a DOI?

Keywords

ResNet; Image Classification; CNN; Data preprocessing; Deep Learning

Abstract

This paper introduces the importance of image classification in computer vision. It aims to classify input images into different categories. Traditional image classification methods use manual feature extraction or feature learning to describe images but it is difficult to reveal deep semantic abstract features. It also requires a lot of manual work. This paper proposes an improved ResNet image classification model that solves problems such as computational complexity and overfitting. The proposed method uses smaller convolutional kernels. Data augmentation techniques are also implemented to improve network performance. By doing so, the algorithm achieves higher accuracy. Results of experiments on CUB200-2011 dataset demonstrate that the improved ResNet model achieves a validation accuracy of 95.50%, significantly outperforming other models. However, some overfitting is observed, indicating the need for further research. The results show the capability of deep learning methods, especially for ResNet model in image classification tasks.

Copyright

© 2023 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Yuheng Wang
PY  - 2023
DA  - 2023/11/27
TI  - Research on Image Classification Based on ResNet
BT  - Proceedings of the 2023 International Conference on Image, Algorithms and Artificial Intelligence (ICIAAI 2023)
PB  - Atlantis Press
SP  - 971
EP  - 979
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-300-9_98
DO  - 10.2991/978-94-6463-300-9_98
ID  - Wang2023
ER  -