Weather Image Recognition Using Vision Transformer

Citation

Tan, Jun Zhi and Lim, Jit Yan and Lim, Kian Ming and Lee, Chin Poo (2023) Weather Image Recognition Using Vision Transformer. In: 2023 IEEE 11th Conference on Systems, Process & Control (ICSPC), 16-16 December 2023, Malacca, Malaysia.

[img] Text
58.pdf - Published Version
Restricted to Repository staff only

Download (413kB)

Abstract

Weather significantly impacts human activities, and accurate weather recognition is crucial to mitigate the risks associated with severe weather conditions. In this research project, we propose Vision Transformer for weather image recognition. The goal is to identify weather patterns and conditions accurately to enhance safety in activities that are affected by the weather. To demonstrate the performance, additional five methods have been adopted to carry out the comparison, including K-Nearest Neighbors, Random Forest, Convolutional Neural Networks, Residual Network, and Compact Convolutional Transformer. Our experimental results show that the proposed Vision Transformer model achieved the highest accuracy of 99.58%, which outperformed the other models. This finding highlights the potential of deep learning techniques for accurate weather recognition

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Weather,
Subjects: Q Science > QC Physics > QC851-999 Meteorology. Climatology Including the earth's atmosphere
Divisions: Faculty of Information Science and Technology (FIST)
Depositing User: Ms Nurul Iqtiani Ahmad
Date Deposited: 27 Mar 2024 03:35
Last Modified: 27 Mar 2024 03:35
URII: http://shdl.mmu.edu.my/id/eprint/12217

Downloads

Downloads per month over past year

View ItemEdit (login required)