Affiliation:
1. School of Computer Science and Technology, Beijing Institute of Technology, Beijing 100081, China
2. School of Information and Electronics, Beijing Institute of Technology, Beijing 100081, China
3. Beijing Institute of Control Engineering, Beijing 100190, China
Abstract
Recently, hyperspectral image classification has made great progress with the development of convolutional neural networks. However, due to the challenges of distribution shifts and data redundancies, the classification accuracy is low. Some existing domain adaptation methods try to mitigate the distribution shifts by training source samples and some labeled target samples. However, in practice, labeled target domain samples are difficult or even impossible to obtain. To solve the above challenges, we propose a novel dual-attention deep discriminative domain generalization framework (DAD3GM) for cross-scene hyperspectral image classification without training the labeled target samples. In DAD3GM, we mainly design two blocks: dual-attention feature learning (DAFL) and deep discriminative feature learning (DDFL). DAFL is designed to extract spatial features by multi-scale self-attention and extract spectral features by multi-head external attention. DDFL is further designed to extract deep discriminative features by contrastive regularization and class discrimination regularization. The combination of DAFL and DDFL can effectively reduce the computational time and improve the generalization performance of DAD3GM. The proposed model achieves 84.25%, 83.53%, and 80.63% overall accuracy on the public Houston, Pavia, and GID benchmarks, respectively. Compared with some classical and state-of-the-art methods, the proposed model achieves optimal results, which reveals its effectiveness and feasibility.
Funder
Pre-research Project on Civil Aerospace Technologies of China National Space Administration
Subject
General Earth and Planetary Sciences
Reference61 articles.
1. Hyperspectral Image Classification—Traditional to Deep Models: A Survey for Future Prospects;Ahmad;IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens.,2022
2. Li, J., Li, X., and Yan, Y. (2023). Unlocking the Potential of Data Augmentation in Contrastive Learning for Hyperspectral Image Classification. Remote Sens., 15.
3. Cross-Domain Contrastive Learning for Hyperspectral Image Classification;Guan;IEEE Trans. Geosci. Remote Sens.,2022
4. Semisupervised Few-Shot Remote Sensing Image Classification Based on KNN Distance Entropy;Chao;IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens.,2022
5. Hyperspectral imagery classification with minimum noise fraction, 2D spatial filtering and SVM;Chen;Int. J. Wavelets Multiresolution Inf. Process.,2022