Cargando…

Classification of Compressed Remote Sensing Multispectral Images via Convolutional Neural Networks

Multispectral sensors constitute a core Earth observation image technology generating massive high-dimensional observations. To address the communication and storage constraints of remote sensing platforms, lossy data compression becomes necessary, but it unavoidably introduces unwanted artifacts. I...

Descripción completa

Detalles Bibliográficos
Autores principales: Giannopoulos, Michalis, Aidini, Anastasia, Pentari, Anastasia, Fotiadou, Konstantina, Tsakalides, Panagiotis
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2020
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8321049/
https://www.ncbi.nlm.nih.gov/pubmed/34460726
http://dx.doi.org/10.3390/jimaging6040024
_version_ 1783730759046004736
author Giannopoulos, Michalis
Aidini, Anastasia
Pentari, Anastasia
Fotiadou, Konstantina
Tsakalides, Panagiotis
author_facet Giannopoulos, Michalis
Aidini, Anastasia
Pentari, Anastasia
Fotiadou, Konstantina
Tsakalides, Panagiotis
author_sort Giannopoulos, Michalis
collection PubMed
description Multispectral sensors constitute a core Earth observation image technology generating massive high-dimensional observations. To address the communication and storage constraints of remote sensing platforms, lossy data compression becomes necessary, but it unavoidably introduces unwanted artifacts. In this work, we consider the encoding of multispectral observations into high-order tensor structures which can naturally capture multi-dimensional dependencies and correlations, and we propose a resource-efficient compression scheme based on quantized low-rank tensor completion. The proposed method is also applicable to the case of missing observations due to environmental conditions, such as cloud cover. To quantify the performance of compression, we consider both typical image quality metrics as well as the impact on state-of-the-art deep learning-based land-cover classification schemes. Experimental analysis on observations from the ESA Sentinel-2 satellite reveals that even minimal compression can have negative effects on classification performance which can be efficiently addressed by our proposed recovery scheme.
format Online
Article
Text
id pubmed-8321049
institution National Center for Biotechnology Information
language English
publishDate 2020
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-83210492021-08-26 Classification of Compressed Remote Sensing Multispectral Images via Convolutional Neural Networks Giannopoulos, Michalis Aidini, Anastasia Pentari, Anastasia Fotiadou, Konstantina Tsakalides, Panagiotis J Imaging Article Multispectral sensors constitute a core Earth observation image technology generating massive high-dimensional observations. To address the communication and storage constraints of remote sensing platforms, lossy data compression becomes necessary, but it unavoidably introduces unwanted artifacts. In this work, we consider the encoding of multispectral observations into high-order tensor structures which can naturally capture multi-dimensional dependencies and correlations, and we propose a resource-efficient compression scheme based on quantized low-rank tensor completion. The proposed method is also applicable to the case of missing observations due to environmental conditions, such as cloud cover. To quantify the performance of compression, we consider both typical image quality metrics as well as the impact on state-of-the-art deep learning-based land-cover classification schemes. Experimental analysis on observations from the ESA Sentinel-2 satellite reveals that even minimal compression can have negative effects on classification performance which can be efficiently addressed by our proposed recovery scheme. MDPI 2020-04-18 /pmc/articles/PMC8321049/ /pubmed/34460726 http://dx.doi.org/10.3390/jimaging6040024 Text en © 2020 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/ (https://creativecommons.org/licenses/by/4.0/) ).
spellingShingle Article
Giannopoulos, Michalis
Aidini, Anastasia
Pentari, Anastasia
Fotiadou, Konstantina
Tsakalides, Panagiotis
Classification of Compressed Remote Sensing Multispectral Images via Convolutional Neural Networks
title Classification of Compressed Remote Sensing Multispectral Images via Convolutional Neural Networks
title_full Classification of Compressed Remote Sensing Multispectral Images via Convolutional Neural Networks
title_fullStr Classification of Compressed Remote Sensing Multispectral Images via Convolutional Neural Networks
title_full_unstemmed Classification of Compressed Remote Sensing Multispectral Images via Convolutional Neural Networks
title_short Classification of Compressed Remote Sensing Multispectral Images via Convolutional Neural Networks
title_sort classification of compressed remote sensing multispectral images via convolutional neural networks
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8321049/
https://www.ncbi.nlm.nih.gov/pubmed/34460726
http://dx.doi.org/10.3390/jimaging6040024
work_keys_str_mv AT giannopoulosmichalis classificationofcompressedremotesensingmultispectralimagesviaconvolutionalneuralnetworks
AT aidinianastasia classificationofcompressedremotesensingmultispectralimagesviaconvolutionalneuralnetworks
AT pentarianastasia classificationofcompressedremotesensingmultispectralimagesviaconvolutionalneuralnetworks
AT fotiadoukonstantina classificationofcompressedremotesensingmultispectralimagesviaconvolutionalneuralnetworks
AT tsakalidespanagiotis classificationofcompressedremotesensingmultispectralimagesviaconvolutionalneuralnetworks