1. Αποθετήριο Παραδοτέων Υποτρόφων Δράσεων ΕΣΠΑ Τριτοβάθμιας Εκπαίδευσης 2017-2023
2. Όλο το Αποθετήριο
3. Ερευνητικές ομάδες

Τύπος:	Άρθρο σε επιστημονικό περιοδικό
Τίτλος:	Classification of compressed remote sensing multispectral images via convolutional neural networks
Συγγραφέας:	[EL] Γιαννόπουλος, Μιχαήλ[EN] Giannopoulos, Michalis [EL] Αϊδίνη, Αναστασία[EN] Aidini, Anastasia [EL] Πεντάρη, Αναστασία[EN] Pentari, Anastasia [EL] Φωτιάδου, Κωνσταντίνα[EN] Fotiadou, Konstantina [EL] Τσακαλίδης, Παναγιώτης[EN] Tsakalides, Panagiotis
Ημερομηνία:	18/04/2020
Περίληψη:	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.
Γλώσσα:	Αγγλικά
Τόπος δημοσίευσης:	Crete, Greece
Σελίδες:	39
DOI:	10.3390/jimaging6040024
ISSN:	2313-433X
Θεματική κατηγορία:	[EL] Μηχανική και συστήματα επικοινωνιών, Τηλεπικοινωνίες[EN] Communication engineering and systems, Telecommunications
Λέξεις-κλειδιά:	multispectral image classification; deep learning; convolutional neural networks; residual learning; compression; quantization; tensor unfoldings; nuclear norm
Κάτοχος πνευματικών δικαιωμάτων:	© 2020 by the authors. Licensee MDPI
Όροι και προϋποθέσεις δικαιωμάτων:	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/).
Ηλεκτρονική διεύθυνση του τεκμηρίου στον εκδότη:	https://www.mdpi.com/2313-433X/6/4/24
Τίτλος πηγής δημοσίευσης:	Journal of Imaging
Τεύχος:	4
Τόμος:	6
Σημειώσεις:	(This article belongs to the Special Issue Multispectral Imaging) Funding: The research work was funded by Greece and the European Union (European Social Fund) in the context of the Youth Employment Initiative through the Operational Program for Human Resources Development, Education and Lifelong Learning, under grant no. MIS 5004457.
Εμφανίζεται στις συλλογές:	Ερευνητικές ομάδες