This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Open AccessArticle
Pyramid Cascaded Convolutional Neural Network with Graph Convolution for Hyperspectral Image Classification
by
Haizhu Pan
Haizhu Pan 1,2,*,
Hui Yan
Hui Yan 1,
Haimiao Ge
Haimiao Ge 1,2
,
Liguo Wang
Liguo Wang
Prof. Dr. Liguo Wang received a BS degree in Bioscience from
the School of Biosciences, Xuzhou He [...]
Prof. Dr. Liguo Wang received a BS degree in Bioscience from
the School of Biosciences, Xuzhou Normal University, Xuzhou, China. He received
a Ph.D degree in Computational Biology from the Beijing Genomics Institute,
Chinese Academy of Sciences, Beijing, China. Prof. Dr. Wang is the consultant
of the Division of Computational Biology, Department of Quantitative Health
Sciences and the Associate Professor of Biomedical Informatics. The research of
Liguo Wang, Ph.D., concentrates on developing computational tools and methods
to transform big data into biological insights and provide cognitive support
for data-informed clinical decision-making and precision medicine. Prof. Dr.
Wang focuses on high-throughput sequencing data analysis, mining and
integration, machine learning, artificial intelligence, and genomic medicine.
3 and
Cuiping Shi
Cuiping Shi 4
1
College of Computer and Control Engineering, Qiqihar University, Qiqihar 161000, China
2
Heilongjiang Key Laboratory of Big Data Network Security Detection and Analysis, Qiqihar University, Qiqihar 161000, China
3
College of Information and Communication Engineering, Dalian Nationalities University, Dalian 116000, China
4
College of Telecommunication and Electronic Engineering, Qiqihar University, Qiqihar 161000, China
*
Author to whom correspondence should be addressed.
Remote Sens. 2024, 16(16), 2942; https://doi.org/10.3390/rs16162942 (registering DOI)
Submission received: 25 June 2024
/
Revised: 4 August 2024
/
Accepted: 8 August 2024
/
Published: 11 August 2024
Abstract
Convolutional neural networks (CNNs) and graph convolutional networks (GCNs) have made considerable advances in hyperspectral image (HSI) classification. However, most CNN-based methods learn features at a single-scale in HSI data, which may be insufficient for multi-scale feature extraction in complex data scenes. To learn the relations among samples in non-grid data, GCNs are employed and combined with CNNs to process HSIs. Nevertheless, most methods based on CNN-GCN may overlook the integration of pixel-wise spectral signatures. In this paper, we propose a pyramid cascaded convolutional neural network with graph convolution (PCCGC) for hyperspectral image classification. It mainly comprises CNN-based and GCN-based subnetworks. Specifically, in the CNN-based subnetwork, a pyramid residual cascaded module and a pyramid convolution cascaded module are employed to extract multiscale spectral and spatial features separately, which can enhance the robustness of the proposed model. Furthermore, an adaptive feature-weighted fusion strategy is utilized to adaptively fuse multiscale spectral and spatial features. In the GCN-based subnetwork, a band selection network (BSNet) is used to learn the spectral signatures in the HSI using nonlinear inter-band dependencies. Then, the spectral-enhanced GCN module is utilized to extract and enhance the important features in the spectral matrix. Subsequently, a mutual-cooperative attention mechanism is constructed to align the spectral signatures between BSNet-based matrix with the spectral-enhanced GCN-based matrix for spectral signature integration. Abundant experiments performed on four widely used real HSI datasets show that our model achieves higher classification accuracy than the fourteen other comparative methods, which shows the superior classification performance of PCCGC over the state-of-the-art methods.
Share and Cite
MDPI and ACS Style
Pan, H.; Yan, H.; Ge, H.; Wang, L.; Shi, C.
Pyramid Cascaded Convolutional Neural Network with Graph Convolution for Hyperspectral Image Classification. Remote Sens. 2024, 16, 2942.
https://doi.org/10.3390/rs16162942
AMA Style
Pan H, Yan H, Ge H, Wang L, Shi C.
Pyramid Cascaded Convolutional Neural Network with Graph Convolution for Hyperspectral Image Classification. Remote Sensing. 2024; 16(16):2942.
https://doi.org/10.3390/rs16162942
Chicago/Turabian Style
Pan, Haizhu, Hui Yan, Haimiao Ge, Liguo Wang, and Cuiping Shi.
2024. "Pyramid Cascaded Convolutional Neural Network with Graph Convolution for Hyperspectral Image Classification" Remote Sensing 16, no. 16: 2942.
https://doi.org/10.3390/rs16162942
Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details
here.
Article Metrics
Article metric data becomes available approximately 24 hours after publication online.