Semantic Segmentation in Satellite Hyperspectral Imagery by Deep Learning

Simple item page
dc.contributor.authorAlvarez Justo, Joncs
dc.contributor.authorGhiţă, Alexandrucs
dc.contributor.authorKováč, Danielcs
dc.contributor.authorL. Garrett, Josephcs
dc.contributor.authorGeorgescu, Mariana-Iulianacs
dc.contributor.authorGonzalez-Llorente, Jesuscs
dc.contributor.authorTudor Ionescu, Raducs
dc.contributor.authorArne Johansen, Torcs
dc.coverage.issue1cs
dc.coverage.volume18cs
dc.date.accessioned2025-04-08T09:56:17Z
dc.date.available2025-04-08T09:56:17Z
dc.date.issued2024-11-07cs
dc.description.abstractSatellites are increasingly adopting onboard AI to optimize operations and increase autonomy through in-orbit inference. The use of deep learning (DL) models for segmentation in hyperspectral imagery offers advantages for remote sensing applications. In this work, we train and test 20 models for multiclass segmentation in hyperspectral imagery, selected for their potential in future space deployment. These models include 1-D and 2-D convolutional neural networks (CNNs) and the latest vision transformers (ViTs). We propose a lightweight 1D-CNN model, 1D-Justo-LiuNet, which outperforms state-of-the-art models in the hypespectral domain. 1D-Justo-LiuNet exceeds the performance of 2D-CNN UNets and outperforms Apple's lightweight vision transformers designed for mobile inference. 1D-Justo-LiuNet achieves the highest accuracy (0.93) with the smallest model size (4563 parameters) among all tested models, while maintaining fast inference. Unlike 2D-CNNs and ViTs, which encode both spectral and spatial information, 1D-Justo-LiuNet focuses solely on the rich spectral features in hyperspectral data, benefitting from the high-dimensional feature space. Our findings are validated across various satellite datasets, with the HYPSO-1 mission serving as the primary case study for sea, land, and cloud segmentation. We further confirm our conclusions through generalization tests on other hyperspectral missions, such as NASA's EO-1. Based on its superior performance and compact size, we conclude that 1D-Justo-LiuNet is highly suitable for in-orbit deployment, providing an effective solution for optimizing and automating satellite operations at edgeen
dc.formattextcs
dc.format.extent273-293cs
dc.format.mimetypeapplication/pdfcs
dc.identifier.citationIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2024, vol. 18, issue 1, p. 273-293.en
dc.identifier.doi10.1109/JSTARS.2024.3487360cs
dc.identifier.issn2151-1535cs
dc.identifier.orcid0000-0003-2701-1802cs
dc.identifier.other193496cs
dc.identifier.scopus57268698100cs
dc.identifier.urihttps://hdl.handle.net/11012/250842
dc.language.isoencs
dc.publisherIEEEcs
dc.relation.ispartofIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensingcs
dc.relation.urihttps://ieeexplore.ieee.org/document/10746584cs
dc.rightsCreative Commons Attribution-NonCommercial-NoDerivatives 4.0 Internationalcs
dc.rights.accessopenAccesscs
dc.rights.sherpahttp://www.sherpa.ac.uk/romeo/issn/2151-1535/cs
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/cs
dc.subject1D-CNNsen
dc.subject2D-CNNsen
dc.subjectdeep learning (DL)en
dc.subjectremote sensingen
dc.subjectsatellite hyperspectral imageryen
dc.subjectsegmentationen
dc.subjectvision transformers (ViTs)en
dc.titleSemantic Segmentation in Satellite Hyperspectral Imagery by Deep Learningen
dc.type.driverarticleen
dc.type.statusPeer-revieweden
dc.type.versionpublishedVersionen
eprints.grantNumberinfo:eu-repo/grantAgreement/TA0/FW/FW09020069cs
sync.item.dbidVAV-193496en
sync.item.dbtypeVAVen
sync.item.insts2025.04.08 11:56:17en
sync.item.modts2025.04.08 11:33:20en
thesis.grantorVysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií. Ústav telekomunikacícs
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Semantic_Segmentation_in_Satellite_Hyperspectral_Imagery_by_Deep_Learning.pdf
Size:
5.48 MB
Format:
Adobe Portable Document Format
Description:
file Semantic_Segmentation_in_Satellite_Hyperspectral_Imagery_by_Deep_Learning.pdf
Download
Collections
Ústav telekomunikací