Triple Graph Convolutional Network for Hyperspectral Image Feature Fusion and Classification

Most graph-based networks utilize superpixel generation methods as a preprocessing step, considering superpixels as graph nodes. In the case of hyperspectral images having high variability in spectral features, considering an image region as a graph node may degrade the class discrimination ability...

Full description

Saved in:
Bibliographic Details
Main Authors: Maryam Imani, Daniele Cerra
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/9/1623
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850278510601437184
author Maryam Imani
Daniele Cerra
author_facet Maryam Imani
Daniele Cerra
author_sort Maryam Imani
collection DOAJ
description Most graph-based networks utilize superpixel generation methods as a preprocessing step, considering superpixels as graph nodes. In the case of hyperspectral images having high variability in spectral features, considering an image region as a graph node may degrade the class discrimination ability of networks for pixel-based classification. Moreover, most graph-based networks focus on global feature extraction, while both local and global information are important for pixel-based classification. To deal with these challenges, superpixel-based graphs are overruled in this work, and a Graph-based Feature Fusion (GF2) method relying on three different graphs is proposed instead. A local patch is considered around each pixel under test, and at the same time, global anchors with the highest informational content are selected from the entire scene. While the first graph explores relationships between neighboring pixels in the local patch and the global anchors, the second and third graphs use the global anchors and pixels of the local patch as nodes, respectively. These graphs are processed using graph convolutional networks, and their results are fused using a cross-attention mechanism. The experiments on three hyperspectral benchmark datasets show that the GF2 network has high classification performance compared to state-of-the-art methods, while imposing a reasonable number of learnable parameters.
format Article
id doaj-art-aeb3e706f5e04b358548e39e734036a8
institution OA Journals
issn 2072-4292
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-aeb3e706f5e04b358548e39e734036a82025-08-20T01:49:28ZengMDPI AGRemote Sensing2072-42922025-05-01179162310.3390/rs17091623Triple Graph Convolutional Network for Hyperspectral Image Feature Fusion and ClassificationMaryam Imani0Daniele Cerra1Faculty of Electrical and Computer Engineering, Tarbiat Modares University, Tehran P.O. Box 14115-111, IranRemote Sensing Technology Institute, German Aerospace Center (DLR), Muenchener Strasse 20, 82234 Wessling, GermanyMost graph-based networks utilize superpixel generation methods as a preprocessing step, considering superpixels as graph nodes. In the case of hyperspectral images having high variability in spectral features, considering an image region as a graph node may degrade the class discrimination ability of networks for pixel-based classification. Moreover, most graph-based networks focus on global feature extraction, while both local and global information are important for pixel-based classification. To deal with these challenges, superpixel-based graphs are overruled in this work, and a Graph-based Feature Fusion (GF2) method relying on three different graphs is proposed instead. A local patch is considered around each pixel under test, and at the same time, global anchors with the highest informational content are selected from the entire scene. While the first graph explores relationships between neighboring pixels in the local patch and the global anchors, the second and third graphs use the global anchors and pixels of the local patch as nodes, respectively. These graphs are processed using graph convolutional networks, and their results are fused using a cross-attention mechanism. The experiments on three hyperspectral benchmark datasets show that the GF2 network has high classification performance compared to state-of-the-art methods, while imposing a reasonable number of learnable parameters.https://www.mdpi.com/2072-4292/17/9/1623graph convolutional networkattention mechanismfeature fusionhyperspectral image classification
spellingShingle Maryam Imani
Daniele Cerra
Triple Graph Convolutional Network for Hyperspectral Image Feature Fusion and Classification
Remote Sensing
graph convolutional network
attention mechanism
feature fusion
hyperspectral image classification
title Triple Graph Convolutional Network for Hyperspectral Image Feature Fusion and Classification
title_full Triple Graph Convolutional Network for Hyperspectral Image Feature Fusion and Classification
title_fullStr Triple Graph Convolutional Network for Hyperspectral Image Feature Fusion and Classification
title_full_unstemmed Triple Graph Convolutional Network for Hyperspectral Image Feature Fusion and Classification
title_short Triple Graph Convolutional Network for Hyperspectral Image Feature Fusion and Classification
title_sort triple graph convolutional network for hyperspectral image feature fusion and classification
topic graph convolutional network
attention mechanism
feature fusion
hyperspectral image classification
url https://www.mdpi.com/2072-4292/17/9/1623
work_keys_str_mv AT maryamimani triplegraphconvolutionalnetworkforhyperspectralimagefeaturefusionandclassification
AT danielecerra triplegraphconvolutionalnetworkforhyperspectralimagefeaturefusionandclassification