Incorporating Attention Mechanism, Dense Connection Blocks, and Multi-Scale Reconstruction Networks for Open-Set Hyperspectral Image Classification
Remote Sensing · 15.09.2023 · 15 (18) · 4535
DOI: 10.3390/rs15184535 ↗Kurz gefasst
Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 15.09.2023. Verfasst von Huaming Zhou und 4 weiteren Autor:innen. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the high-end foreign expert introduction program, Heilongjiang Natural Science Foundation Project, Reserved Leaders of Heilongjiang Provincial Leading Talent Echelon (2021).
Abstract (Originalsprache)
Hyperspectral image classification plays a crucial role in various remote sensing applications. However, existing methods often struggle with the challenge of unknown classes, leading to decreased classification accuracy and limited generalization. In this paper, we propose a novel deep learning framework called IADMRN, which addresses the issue of unknown class handling in hyperspectral image classification. IADMRN combines the strengths of dense connection blocks and attention mechanisms to extract discriminative features from hyperspectral data. Furthermore, it employs a multi-scale deconvolution image reconstruction sub-network to enhance feature reconstruction and provide additional information for classification. To handle unknown classes, IADMRN utilizes an extreme value theory-based model to calculate the probability of unknown class membership. Experimental results on the three public datasets demonstrate that IADMRN outperforms state-of-the-art methods in terms of classification accuracy for both known and unknown classes. Experimental results show that the proposed methods outperform several state-of-the-art methods, which outperformed DCFSL by 8.47%, 6.57%, and 4.25%, and outperformed MDL4OW by 4.35%, 4.08%, and 2.47% on the Salinas, University of Pavia, and Indian Pines datasets, respectively. The proposed framework is computationally efficient and showcases the ability to effectively handle unknown classes in hyperspectral image classification tasks. Overall, IADMRN offers a promising solution for accurate and robust hyperspectral image classification, making it a valuable tool for remote sensing applications.
Förderung
- the high-end foreign expert introduction program
- Heilongjiang Natural Science Foundation Project
- Reserved Leaders of Heilongjiang Provincial Leading Talent Echelon (2021)
Zitieren
APA
Zhou, H., Wu, H., Wang, A., Iwahori, Y., & Yu, X. (2023). Incorporating Attention Mechanism, Dense Connection Blocks, and Multi-Scale Reconstruction Networks for Open-Set Hyperspectral Image Classification. Remote Sensing, 15(18), 4535. https://doi.org/10.3390/rs15184535
Vancouver
Zhou H, Wu H, Wang A, Iwahori Y, Yu X. Incorporating Attention Mechanism, Dense Connection Blocks, and Multi-Scale Reconstruction Networks for Open-Set Hyperspectral Image Classification. Remote Sensing. 2023;15(18):4535. doi:10.3390/rs15184535
BibTeX
@article{Zhou2023,
author = {Zhou, Huaming and Wu, Haibin and Wang, Aili and Iwahori, Yuji and Yu, Xiaoyu},
title = {Incorporating Attention Mechanism, Dense Connection Blocks, and Multi-Scale Reconstruction Networks for Open-Set Hyperspectral Image Classification},
journal = {Remote Sensing},
year = {2023},
volume = {15},
number = {18},
pages = {4535},
doi = {10.3390/rs15184535}
}
EUVIMED