Novel Gumbel-Softmax trick enabled concrete autoencoder with entropy constraints for unsupervised hyperspectral band selection

Simple item page
dc.contributor.authorSun, He
dc.contributor.authorRen, Jinchang
dc.contributor.authorZhao, Huimin
dc.contributor.authorYuen, Peter W. T.
dc.contributor.authorTschannerl, Julius
dc.date.accessioned2021-07-13T14:39:03Z
dc.date.available2021-07-13T14:39:03Z
dc.date.issued2021-06-04
dc.description.abstractAs an important topic in hyperspectral image (HSI) analysis, band selection has attracted increasing attention in the last two decades for dimensionality reduction in HSI. With the great success of deep learning (DL)-based models recently, a robust unsupervised band selection (UBS) neural network is highly desired, particularly due to the lack of sufficient ground truth information to train the DL networks. Existing DL models for band selection either depend on the class label information or have unstable results via ranking the learned weights. To tackle these challenging issues, in this article, we propose a Gumbel-Softmax (GS) trick enabled concrete autoencoder-based UBS framework (CAE-UBS) for HSI, in which the learning process is featured by the introduced concrete random variables and the reconstruction loss. By searching from the generated potential band selection candidates from the concrete encoder, the optimal band subset can be selected based on an information entropy (IE) criterion. The idea of the CAE-UBS is quite straightforward, which does not rely on any complicated strategies or metrics. The robust performance on four publicly available datasets has validated the superiority of our CAE-UBS framework in the classification of the HSIs.en_UK
dc.identifier.citationSun H, Ren J, Zhao H, et al., (2021) Novel Gumbel-Softmax trick enabled concrete autoencoder with entropy constraints for unsupervised hyperspectral band selection. IEEE Transactions on Geoscience and Remote Sensing, Volume 60, June 2021 Article number 5506413en_UK
dc.identifier.issn0196-2892
dc.identifier.urihttps://doi.org/10.1109/TGRS.2021.3075663
dc.identifier.urihttp://dspace.lib.cranfield.ac.uk/handle/1826/16877
dc.language.isoenen_UK
dc.publisherIEEEen_UK
dc.rightsAttribution-NonCommercial 4.0 International*
dc.rights.urihttp://creativecommons.org/licenses/by-nc/4.0/*
dc.subjectunsupervised band selection (UBS)en_UK
dc.subjectinformation entropy (IE)en_UK
dc.subjecthyperspectral image (HSI)en_UK
dc.subjectconcrete random variableen_UK
dc.subjectAutoencoder (AE)en_UK
dc.titleNovel Gumbel-Softmax trick enabled concrete autoencoder with entropy constraints for unsupervised hyperspectral band selectionen_UK
dc.typeArticleen_UK

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Novel_Gumbel-Softmax_trick_enabled_concrete_autoencoder_entropy_constraints-2021.pdf
Size:
2.75 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.63 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Staff publications (CDS)