He Sun
Novel hyperbolic clustering-based band hierarchy (HCBH) for effective unsupervised band selection of hyperspectral images.
Sun, He; Zhang, Lei; Ren, Jinchang; Huang, Hua
Authors
Lei Zhang
Jinchang Ren
Hua Huang
Abstract
For dimensionality reduction of HSI, many clustering-based unsupervised band selection (UBS) methods have been proposed due to their superiority of reducing the high redundancy between selected bands. However, most of these methods fail to reflect the data structure of HSI, leading to inconsistent results of band selection. To tackle this particular issue, we have proposed a novel hyperbolic clustering-based band hierarchy (HCBH) to fully represent the underlying spectral structure and obtain a more consistent band selection. With the proposed adaptive hyperbolic clustering, the performance can be effectively improved with the aid of geometrical information. By introducing a cluster-centre based ranking metric, the desired band subset can be naturally obtained during the clustering process. Experimental results on three popularly used datasets have validated the superior performance of the proposed approach, which outperforms a few state-of-the-art (SOTA) UBS approaches.
Citation
SUN, H., ZHANG, L., REN, J. and HUANG, H. 2022. Novel hyperbolic clustering-based band hierarchy (HCBH) for effective unsupervised band selection of hyperspectral images. Pattern recognition [online], 130, article number 108788. Available from: https://doi.org/10.1016/j.patcog.2022.108788
Journal Article Type | Article |
---|---|
Acceptance Date | May 11, 2022 |
Online Publication Date | May 13, 2022 |
Publication Date | Oct 31, 2022 |
Deposit Date | Jun 30, 2022 |
Publicly Available Date | May 14, 2023 |
Journal | Pattern Recognition |
Print ISSN | 0031-3203 |
Publisher | Elsevier |
Peer Reviewed | Peer Reviewed |
Volume | 130 |
Article Number | 108788 |
DOI | https://doi.org/10.1016/j.patcog.2022.108788 |
Keywords | Hyperspectral images; Unsupervised band selection; Hyperbolic space clustering; Hierarchical clustering |
Public URL | https://rgu-repository.worktribe.com/output/1670295 |
Files
This file is under embargo until May 14, 2023 due to copyright reasons.
Contact publications@rgu.ac.uk to request a copy for personal use.