Opin vísindi

Blind Hyperspectral Unmixing using Autoencoders : A Critical Comparison

Show simple item record

dc.contributor.author Pálsson, Burkni
dc.contributor.author Sveinsson, Jóhannes Rúnar
dc.contributor.author Úlfarsson, Magnús Örn
dc.date.accessioned 2022-04-12T01:02:21Z
dc.date.available 2022-04-12T01:02:21Z
dc.date.issued 2022-01-06
dc.identifier.citation Pálsson , B , Sveinsson , J R & Úlfarsson , M Ö 2022 , ' Blind Hyperspectral Unmixing using Autoencoders : A Critical Comparison ' , IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , vol. 15 , pp. 1340-1372 . https://doi.org/10.1109/JSTARS.2021.3140154
dc.identifier.issn 1939-1404
dc.identifier.other PURE: 45117707
dc.identifier.other PURE UUID: 279bd3e4-cec0-4358-8d10-411c228ceee2
dc.identifier.other Scopus: 85122884562
dc.identifier.uri https://hdl.handle.net/20.500.11815/3036
dc.description Publisher Copyright: The Authors
dc.description.abstract Deep learning has shown to be a powerful tool and has heavily impacted the data-intensive field of remote sensing. As a result, the number of published deep learning-based spectral unmixing techniques is proliferating. Blind hyperspectral unmixing (HU) is the process of resolving the measured spectrum of a pixel into a combination of a set of spectral signatures called endmembers and simultaneously determining their fractional abundances in the pixel. This paper details the various autoencoder architectures used in HU and provides a critical comparison of some of the existing published blind unmixing methods based on autoencoders. Eleven different autoencoder methods and one traditional method will be compared in blind unmixing experiments using four real datasets and four synthetic datasets with different spectral variability. Additionally, extensive ablation experiments with a simple spectral unmixing autoencoder will be performed. The results are interpreted in terms of the various implementation details, and the question of why autoencoder methods are so powerful compared to traditional methods is unraveled. The source codes for all methods implemented in this paper can be found at the following location: https://github.com/burknipalsson/hu_autoencoders.
dc.format.extent 32
dc.format.extent 1340-1372
dc.language.iso en
dc.relation.ispartofseries IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing; 15()
dc.rights info:eu-repo/semantics/openAccess
dc.subject Myndvinnsla
dc.subject Fjarkönnun
dc.subject Litrófsgreining
dc.subject Líkanagerð
dc.subject Atmospheric modeling
dc.subject autoencoder
dc.subject deep learning
dc.subject Feature extraction
dc.subject Hyperspectral data unmixing
dc.subject Hyperspectral imaging
dc.subject image processing
dc.subject multitask learning
dc.subject neural network
dc.subject Object detection
dc.subject Sparse matrices
dc.subject Spatial resolution
dc.subject spectral-spatial model
dc.subject Spectroscopy
dc.subject Computers in Earth Sciences
dc.subject Atmospheric Science
dc.title Blind Hyperspectral Unmixing using Autoencoders : A Critical Comparison
dc.type /dk/atira/pure/researchoutput/researchoutputtypes/contributiontojournal/article
dc.description.version Peer reviewed
dc.identifier.doi https://doi.org/10.1109/JSTARS.2021.3140154
dc.relation.url http://www.scopus.com/inward/record.url?scp=85122884562&partnerID=8YFLogxK
dc.contributor.department Faculty of Electrical and Computer Engineering


Files in this item

This item appears in the following Collection(s)

Show simple item record