Multiscale Feature Fusion for Hyperspectral Image Classification Using Hybrid 3D-2D Depthwise Separable Convolution Networks

Basit Öğe Kaydı
dc.authoridFIRAT, Huseyin/0000-0002-1257-8518
dc.authoridHanbay, Davut/0000-0003-2271-7865
dc.authoridÇİĞ, HARUN/0000-0003-0419-9531
dc.authorwosidFIRAT, Huseyin/ABB-7417-2021
dc.authorwosidHanbay, Davut/AAG-8511-2019
dc.authorwosidÇİĞ, HARUN/ABA-3476-2020
dc.contributor.authorFirat, Hueseyin
dc.contributor.authorCig, Harun
dc.contributor.authorGuellueoglu, Mehmet Tahir
dc.contributor.authorAsker, Mehmet Emin
dc.contributor.authorHanbay, Davut
dc.date.accessioned2024-08-04T20:54:51Z
dc.date.available2024-08-04T20:54:51Z
dc.date.issued2023
dc.departmentİnönü Üniversitesien_US
dc.description.abstractHyperspectral remote sensing images (HRSI) comprise three-dimensional image cubes, containing a single spectral dimension alongside two spatial dimensions. HRSI are presently among the foremost essential datasets for Earth observation. The task of HRSI classification is intricate due to the influence of spectral mixing, leading to notable variability within classes and resemblances across classes. Consequently, the field of HRSI classification has garnered significant research attention in recent times. Convolutional Neural Networks (CNNs) are harnessed to address these issues, enabling both feature extraction and classification. This study introduces a novel approach for HRSI classification called the hybrid 3D-2D depthwise separable convolution network (Hybrid DSCNet), which leverages multiscale feature integration. Within the Hybrid DSCNet, diverse kernel sizes contribute to an enriched feature extraction process from HRSI. The conventional 3D-2D CNN, while effective, comes with a computational load. Instead of using the standard 3D-2D CNN, this study adopts the 3D-2D DSC architecture. This approach partitions the conventional convolution into two components: pointwise and depthwise convolution, yielding a substantial reduction in trainable parameters and computational complexity. To evaluate the proposed method, the Indian Pines dataset along with WHU-Hi subdatasets (LongKou-LK, HanChuan-HC, and HongHu-HH) were employed. Employing a 5% training sample, impressive overall accuracy scores were achieved: 94.51%, 99.78%, 97.06%, and 97.27% for Indian Pines, WHU-LK, WHU-HC, and WHU-HH, respectively. Comparative analysis of the proposed approach with cutting-edge techniques within the literature reveals its superior performance across the four HRSI datasets. Notably, the Hybrid DSCNet attains enhanced classification accuracy while maintaining lower computational overhead.en_US
dc.identifier.doi10.18280/ts.400512
dc.identifier.endpage1939en_US
dc.identifier.issn0765-0019
dc.identifier.issn1958-5608
dc.identifier.issue5en_US
dc.identifier.scopus2-s2.0-85177879498en_US
dc.identifier.scopusqualityQ3en_US
dc.identifier.startpage1921en_US
dc.identifier.urihttps://doi.org/10.18280/ts.400512
dc.identifier.urihttps://hdl.handle.net/11616/101686
dc.identifier.volume40en_US
dc.identifier.wosWOS:001094288100012en_US
dc.identifier.wosqualityQ4en_US
dc.indekslendigikaynakWeb of Scienceen_US
dc.indekslendigikaynakScopusen_US
dc.language.isoenen_US
dc.publisherInt Information & Engineering Technology Assocen_US
dc.relation.ispartofTraitement Du Signalen_US
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanıen_US
dc.rightsinfo:eu-repo/semantics/openAccessen_US
dc.subjectdepthwise separable convolution (DSC),en_US
dc.subjectconvolutional neural network (CNN),en_US
dc.subjecthyperspectral image classification remoteen_US
dc.subjectsensing hybrid CNNen_US
dc.titleMultiscale Feature Fusion for Hyperspectral Image Classification Using Hybrid 3D-2D Depthwise Separable Convolution Networksen_US
dc.typeArticleen_US

Dosyalar

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu