Onboard Processing of Hyperspectral Imagery: Deep Learning Advancements, Methodologies, Challenges, and Emerging Trends
Recent advancements in deep learning techniques have spurred considerable interest in their application to hyperspectral imagery processing. This article provides a comprehensive review of the latest developments in this field, focusing on methodologies, challenges, and emerging trends. Deep learnin...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10834581/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1825207016632614912 |
|---|---|
| author | Nafiseh Ghasemi Jon Alvarez Justo Marco Celesti Laurent Despoisse Jens Nieke |
| author_facet | Nafiseh Ghasemi Jon Alvarez Justo Marco Celesti Laurent Despoisse Jens Nieke |
| author_sort | Nafiseh Ghasemi |
| collection | DOAJ |
| description | Recent advancements in deep learning techniques have spurred considerable interest in their application to hyperspectral imagery processing. This article provides a comprehensive review of the latest developments in this field, focusing on methodologies, challenges, and emerging trends. Deep learning architectures, such as convolutional neural networks (CNNs), autoencoders, deep belief networks, generative adverserial networks (GANs), and recurrent neural networks are examined for their suitability in processing hyperspectral data. Key challenges, including limited training data and computational constraints, are identified, along with strategies, such as data augmentation and noise reduction using GANs. This article discusses the efficacy of different network architectures, highlighting the advantages of lightweight CNN models and 1D-CNNs for onboard processing. Moreover, the potential of hardware accelerators, particularly field programmable gate arrays, for enhancing processing efficiency is explored. This review concludes with insights into ongoing research trends, including the integration of deep learning techniques into Earth observation missions, such as the Copernicus hyperspectral imaging mission for the environment mission, and emphasizes the need for further exploration and refinement of deep learning methodologies to address the evolving demands of hyperspectral image processing. |
| format | Article |
| id | doaj-art-512b565506ef418ba5d8f2d944eabb9a |
| institution | Kabale University |
| issn | 1939-1404 2151-1535 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj-art-512b565506ef418ba5d8f2d944eabb9a2025-02-07T00:00:25ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-01184780479010.1109/JSTARS.2025.352789810834581Onboard Processing of Hyperspectral Imagery: Deep Learning Advancements, Methodologies, Challenges, and Emerging TrendsNafiseh Ghasemi0https://orcid.org/0009-0002-2699-0279Jon Alvarez Justo1https://orcid.org/0000-0003-4867-2693Marco Celesti2Laurent Despoisse3Jens Nieke4European Space Research and Technology Centre (ESTEC), Noordwijk, AZ, NetherlandsDepartment of Engineering Cybernetics, Norwegian University of Science and Technology (NTNU), Trondheim, NorwayEuropean Space Research and Technology Centre (ESTEC), Noordwijk, AZ, NetherlandsEuropean Space Research and Technology Centre (ESTEC), Noordwijk, AZ, NetherlandsEuropean Space Research and Technology Centre (ESTEC), Noordwijk, AZ, NetherlandsRecent advancements in deep learning techniques have spurred considerable interest in their application to hyperspectral imagery processing. This article provides a comprehensive review of the latest developments in this field, focusing on methodologies, challenges, and emerging trends. Deep learning architectures, such as convolutional neural networks (CNNs), autoencoders, deep belief networks, generative adverserial networks (GANs), and recurrent neural networks are examined for their suitability in processing hyperspectral data. Key challenges, including limited training data and computational constraints, are identified, along with strategies, such as data augmentation and noise reduction using GANs. This article discusses the efficacy of different network architectures, highlighting the advantages of lightweight CNN models and 1D-CNNs for onboard processing. Moreover, the potential of hardware accelerators, particularly field programmable gate arrays, for enhancing processing efficiency is explored. This review concludes with insights into ongoing research trends, including the integration of deep learning techniques into Earth observation missions, such as the Copernicus hyperspectral imaging mission for the environment mission, and emphasizes the need for further exploration and refinement of deep learning methodologies to address the evolving demands of hyperspectral image processing.https://ieeexplore.ieee.org/document/10834581/ClassificationCopernicus hyperspectral imaging mission for the environment (CHIME) missiondeep learning (DL)hardware acceleratorshyperspectral (HS)image processing |
| spellingShingle | Nafiseh Ghasemi Jon Alvarez Justo Marco Celesti Laurent Despoisse Jens Nieke Onboard Processing of Hyperspectral Imagery: Deep Learning Advancements, Methodologies, Challenges, and Emerging Trends IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Classification Copernicus hyperspectral imaging mission for the environment (CHIME) mission deep learning (DL) hardware accelerators hyperspectral (HS) image processing |
| title | Onboard Processing of Hyperspectral Imagery: Deep Learning Advancements, Methodologies, Challenges, and Emerging Trends |
| title_full | Onboard Processing of Hyperspectral Imagery: Deep Learning Advancements, Methodologies, Challenges, and Emerging Trends |
| title_fullStr | Onboard Processing of Hyperspectral Imagery: Deep Learning Advancements, Methodologies, Challenges, and Emerging Trends |
| title_full_unstemmed | Onboard Processing of Hyperspectral Imagery: Deep Learning Advancements, Methodologies, Challenges, and Emerging Trends |
| title_short | Onboard Processing of Hyperspectral Imagery: Deep Learning Advancements, Methodologies, Challenges, and Emerging Trends |
| title_sort | onboard processing of hyperspectral imagery deep learning advancements methodologies challenges and emerging trends |
| topic | Classification Copernicus hyperspectral imaging mission for the environment (CHIME) mission deep learning (DL) hardware accelerators hyperspectral (HS) image processing |
| url | https://ieeexplore.ieee.org/document/10834581/ |
| work_keys_str_mv | AT nafisehghasemi onboardprocessingofhyperspectralimagerydeeplearningadvancementsmethodologieschallengesandemergingtrends AT jonalvarezjusto onboardprocessingofhyperspectralimagerydeeplearningadvancementsmethodologieschallengesandemergingtrends AT marcocelesti onboardprocessingofhyperspectralimagerydeeplearningadvancementsmethodologieschallengesandemergingtrends AT laurentdespoisse onboardprocessingofhyperspectralimagerydeeplearningadvancementsmethodologieschallengesandemergingtrends AT jensnieke onboardprocessingofhyperspectralimagerydeeplearningadvancementsmethodologieschallengesandemergingtrends |