Efficient Unsupervised Clustering of Hyperspectral Images via Flexible Multi-Anchor Graphs

Unsupervised hyperspectral image (HSI) clustering is a fundamental yet challenging task due to high dimensionality and complex spectral–spatial characteristics. In this paper, we propose a novel and efficient clustering framework centered on adaptive and diverse anchor graph modeling. First, we intr...

Full description

Saved in:
Bibliographic Details
Main Authors: Yihong Li, Ting Wang, Zhe Cao, Haonan Xin, Rong Wang
Format: Article
Language:English
Published: MDPI AG 2025-07-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/15/2647
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849765947406024704
author Yihong Li
Ting Wang
Zhe Cao
Haonan Xin
Rong Wang
author_facet Yihong Li
Ting Wang
Zhe Cao
Haonan Xin
Rong Wang
author_sort Yihong Li
collection DOAJ
description Unsupervised hyperspectral image (HSI) clustering is a fundamental yet challenging task due to high dimensionality and complex spectral–spatial characteristics. In this paper, we propose a novel and efficient clustering framework centered on adaptive and diverse anchor graph modeling. First, we introduce a parameter-free construction strategy that employs Entropy Rate Superpixel (ERS) segmentation to generate multiple anchor graphs of varying sizes from a single HSI, overcoming the limitation of fixed anchor quantities and enhancing structural expressiveness. Second, we propose an anchor-to-pixel label propagation mechanism to transfer anchor-level cluster labels back to the pixel level, reinforcing spatial coherence and spectral discriminability. Third, we perform clustering directly at the anchor level, which substantially reduces computational cost while retaining structure-aware accuracy. Extensive experiments on three benchmark datasets (Trento, Salinas, and Pavia Center) demonstrate the effectiveness and efficiency of our approach.
format Article
id doaj-art-5de4cd0cdfd745c1b06f82f15a1fed8f
institution DOAJ
issn 2072-4292
language English
publishDate 2025-07-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-5de4cd0cdfd745c1b06f82f15a1fed8f2025-08-20T03:04:43ZengMDPI AGRemote Sensing2072-42922025-07-011715264710.3390/rs17152647Efficient Unsupervised Clustering of Hyperspectral Images via Flexible Multi-Anchor GraphsYihong Li0Ting Wang1Zhe Cao2Haonan Xin3Rong Wang4Rocket Force University of Engineering, Xi’an 710025, ChinaThe School of Artificial Intelligence, Optics and Electronics (iOPEN), Northwestern Polytechnical University, Xi’an 710072, ChinaThe School of Artificial Intelligence, Optics and Electronics (iOPEN), Northwestern Polytechnical University, Xi’an 710072, ChinaThe School of Artificial Intelligence, Optics and Electronics (iOPEN), Northwestern Polytechnical University, Xi’an 710072, ChinaThe School of Artificial Intelligence, Optics and Electronics (iOPEN), Northwestern Polytechnical University, Xi’an 710072, ChinaUnsupervised hyperspectral image (HSI) clustering is a fundamental yet challenging task due to high dimensionality and complex spectral–spatial characteristics. In this paper, we propose a novel and efficient clustering framework centered on adaptive and diverse anchor graph modeling. First, we introduce a parameter-free construction strategy that employs Entropy Rate Superpixel (ERS) segmentation to generate multiple anchor graphs of varying sizes from a single HSI, overcoming the limitation of fixed anchor quantities and enhancing structural expressiveness. Second, we propose an anchor-to-pixel label propagation mechanism to transfer anchor-level cluster labels back to the pixel level, reinforcing spatial coherence and spectral discriminability. Third, we perform clustering directly at the anchor level, which substantially reduces computational cost while retaining structure-aware accuracy. Extensive experiments on three benchmark datasets (Trento, Salinas, and Pavia Center) demonstrate the effectiveness and efficiency of our approach.https://www.mdpi.com/2072-4292/17/15/2647hyperspectral image clusteringanchor graph modelingunsupervised learningERS segmentationlabel propagationefficient clustering
spellingShingle Yihong Li
Ting Wang
Zhe Cao
Haonan Xin
Rong Wang
Efficient Unsupervised Clustering of Hyperspectral Images via Flexible Multi-Anchor Graphs
Remote Sensing
hyperspectral image clustering
anchor graph modeling
unsupervised learning
ERS segmentation
label propagation
efficient clustering
title Efficient Unsupervised Clustering of Hyperspectral Images via Flexible Multi-Anchor Graphs
title_full Efficient Unsupervised Clustering of Hyperspectral Images via Flexible Multi-Anchor Graphs
title_fullStr Efficient Unsupervised Clustering of Hyperspectral Images via Flexible Multi-Anchor Graphs
title_full_unstemmed Efficient Unsupervised Clustering of Hyperspectral Images via Flexible Multi-Anchor Graphs
title_short Efficient Unsupervised Clustering of Hyperspectral Images via Flexible Multi-Anchor Graphs
title_sort efficient unsupervised clustering of hyperspectral images via flexible multi anchor graphs
topic hyperspectral image clustering
anchor graph modeling
unsupervised learning
ERS segmentation
label propagation
efficient clustering
url https://www.mdpi.com/2072-4292/17/15/2647
work_keys_str_mv AT yihongli efficientunsupervisedclusteringofhyperspectralimagesviaflexiblemultianchorgraphs
AT tingwang efficientunsupervisedclusteringofhyperspectralimagesviaflexiblemultianchorgraphs
AT zhecao efficientunsupervisedclusteringofhyperspectralimagesviaflexiblemultianchorgraphs
AT haonanxin efficientunsupervisedclusteringofhyperspectralimagesviaflexiblemultianchorgraphs
AT rongwang efficientunsupervisedclusteringofhyperspectralimagesviaflexiblemultianchorgraphs