This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Open AccessReview
Generative Adversarial Networks for SAR Automatic Target Recognition and Classification Models Enhanced Explainability: Perspectives and Challenges
by
Héloïse Remusati
Héloïse Remusati 1,*,†
,
Jean-Marc Le Caillec
Jean-Marc Le Caillec
Prof. Jean-Marc Le Caillec is a professor in the Department of Information and Image Processing at a [...]
Prof. Jean-Marc Le Caillec is a professor in the Department of Information and Image Processing at IMT Atlantique (Ecole Nationale Superieure des Télécommunications de Bretagne, Telecom Bretagne). He obtained the degree of Engineer in Telecommunications in 1992 from Telecom Bretagne. He received a Ph.D. in Mathematics and Signal Processing from the University of Rennes I in 1992. From 1997 to 1999, he worked for Thomson AirSys (now Thales AirSys). He joined Telecom Bretagne as an associate professor in 1999 and became a full professor in 2007. From 2014 to 2020, he was in charge of the division CID (Knowledge, Information, Decision) of the Lab-STICC (Laboratory of Sciences and Technics for Information, Communication and Knowledge). He is also an associate professor at Laval University (Québec, Canada). He represented France in the former NATO group SET-208 on the signal fusion for ground-penetrating radars. He is a member of the scientific committee of LATERAL (joint lab between Lab-Sticc and Thales LAS) and CORMORAN (joint lab between Lab-Sticc and Thales DMS).
2,†
,
Jean-Yves Schneider
Jean-Yves Schneider 1,
Jacques Petit-Frère
Jacques Petit-Frère 1 and
Thomas Merlet
Thomas Merlet
Dr. Thomas Merlet received his M.Sc. in Engineering from the Ecole Supérieure de Physique et Chimie [...]
Dr. Thomas Merlet received his M.Sc. in Engineering from the Ecole Supérieure de Physique et Chimie Industrielle, Paris, France, in 1993, and received his Ph.D. in opto-RF Components from the University of Paris XI, Orsay, France, in 1997. He is currently the Head of the Missile Electronics Sector of the Joint Laboratory LATERAL (Lab-STICC Thales Research Alliance) and managing future products and innovation activities of the Electronics Missiles Division, Thales, Elancourt, France.
1
1
Thales LAS France SAS, 2 Avenue Gay Lussac, 78990 Elancourt, France
2
IMT Atlantique Bretagne-Pays de la Loire, Campus de Brest, Technopole Brest-Iroise, CS 83818, CEDEX 3, 29238 Brest, France
*
Author to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Remote Sens. 2024, 16(14), 2569; https://doi.org/10.3390/rs16142569 (registering DOI)
Submission received: 4 June 2024
/
Revised: 3 July 2024
/
Accepted: 8 July 2024
/
Published: 13 July 2024
Abstract
Generative adversarial networks (or GANs) are a specific deep learning architecture often used for different usages, such as data generation or image-to-image translation. In recent years, this structure has gained increased popularity and has been used in different fields. One area of expertise currently in vogue is the use of GANs to produce synthetic aperture radar (SAR) data, and especially expand training datasets for SAR automatic target recognition (ATR). In effect, the complex SAR image formation makes these kind of data rich in information, leading to the use of deep networks in deep learning-based methods. Yet, deep networks also require sufficient data for training. However, contrary to optical images, we generally do not have a substantial number of available SAR images because of their acquisition and labelling cost; GANs are then an interesting tool. Concurrently, how to improve explainability for SAR ATR deep neural networks and how to make their reasoning more transparent have been increasingly explored as model opacity deteriorates trust of users. This paper aims at reviewing how GANs are used with SAR images, but also giving perspectives on how GANs could be used to improve interpretability and explainability of SAR classifiers.
Share and Cite
MDPI and ACS Style
Remusati, H.; Le Caillec, J.-M.; Schneider, J.-Y.; Petit-Frère, J.; Merlet, T.
Generative Adversarial Networks for SAR Automatic Target Recognition and Classification Models Enhanced Explainability: Perspectives and Challenges. Remote Sens. 2024, 16, 2569.
https://doi.org/10.3390/rs16142569
AMA Style
Remusati H, Le Caillec J-M, Schneider J-Y, Petit-Frère J, Merlet T.
Generative Adversarial Networks for SAR Automatic Target Recognition and Classification Models Enhanced Explainability: Perspectives and Challenges. Remote Sensing. 2024; 16(14):2569.
https://doi.org/10.3390/rs16142569
Chicago/Turabian Style
Remusati, Héloïse, Jean-Marc Le Caillec, Jean-Yves Schneider, Jacques Petit-Frère, and Thomas Merlet.
2024. "Generative Adversarial Networks for SAR Automatic Target Recognition and Classification Models Enhanced Explainability: Perspectives and Challenges" Remote Sensing 16, no. 14: 2569.
https://doi.org/10.3390/rs16142569
Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details
here.
Article Metrics
Article metric data becomes available approximately 24 hours after publication online.