Affiliation:
1. College of Electrical Engineering, Zhejiang University, Hangzhou 310027, China
Abstract
Despite the rapid development of deep learning in hyperspectral image classification (HSIC), most models require a large amount of labeled data, which are both time-consuming and laborious to obtain. However, contrastive learning can extract spatial–spectral features from samples without labels, which helps to solve the above problem. Our focus is on optimizing the contrastive learning process and improving feature extraction from all samples. In this study, we propose the Unlocking-the-Potential-of-Data-Augmentation (UPDA) strategy, which involves adding superior data augmentation methods to enhance the representation of features extracted by contrastive learning. Specifically, we introduce three augmentation methods—band erasure, gradient mask, and random occlusion—to the Bootstrap-Your-Own-Latent (BYOL) structure. Our experimental results demonstrate that our method can effectively improve feature representation and thus improve classification accuracy. Additionally, we conduct ablation experiments to explore the effectiveness of different data augmentation methods.
Funder
National Nature Science Foundation of China
Joint Fund Project of Chinese Ministry of Education
Subject
General Earth and Planetary Sciences
Reference41 articles.
1. Hyperspectral image classification: Potentials, challenges, and future directions;Datta;Comput. Intell. Neurosci.,2022
2. Stuart, M.B., McGonigle, A.J., and Willmott, J.R. (2019). Hyperspectral imaging in environmental monitoring: A review of recent developments and technological advances in compact field deployable systems. Sensors, 19.
3. Urban land cover classification with airborne hyperspectral data: What features to use?;Tong;IEEE J. Sel. Top. Appl. Earth Obs. Remote. Sens.,2013
4. Fusion of Dual Spatial Information for Hyperspectral Image Classification;Duan;IEEE Trans. Geosci. Remote. Sens.,2021
5. Principal component analysis;Wold;Chemom. Intell. Lab. Syst.,1987
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献