GPU-friendly neural networks for remote sensing scene classification

Haut, Juan M. ; Alcolea, Adrian (Universidad de Zaragoza) ; Paoletti, Mercedes E. ; Plaza, Javier ; Resano, Javier (Universidad de Zaragoza) ; Plaza, Antonio

Resumen: Convolutional neural networks (CNNs) have proven to be very efficient for the analysis of remote sensing (RS) images. Due to the inherent complexity of extracting features from these images, along with the increasing amount of data to be processed (and the diversity of applications), there is a clear tendency to develop and employ increasingly deep and complex CNNs. In this regard, graphics processing units (GPUs) are frequently used to optimize their execution, both for the training and inference stages, optimizing the performance of neural models through their many-core architecture. Hence, the efficient use of the GPU resources should be at the core of optimizations. This letter analyzes the possibilities of using a new family of CNNs, denoted as TResNets, to provide an efficient solution to the RS scene classification problem. Moreover, the considered models have been combined with mixed precision to enhance their training performance. Our experimental results, conducted over three publicly available RS data sets, show that the proposed networks achieve better accuracy and more efficient use of GPU resources than other state-of-the-art networks. Source code is available at https://github.com/mhaut/GPUfriendlyRS.

Idioma: Inglés

DOI: 10.1109/LGRS.2020.3019378

Año: 2022

Publicado en: IEEE Geoscience and Remote Sensing Letters 19 (2022), 8001005 [5 pp.]
ISSN: 1545-598X

Factor impacto JCR: 4.8 (2022)

Categ. JCR: GEOCHEMISTRY & GEOPHYSICS rank: 10 / 87 = 0.115 (2022) - Q1 - T1
Categ. JCR: IMAGING SCIENCE & PHOTOGRAPHIC TECHNOLOGY rank: 9 / 28 = 0.321 (2022) - Q2 - T1
Categ. JCR: REMOTE SENSING rank: 14 / 34 = 0.412 (2022) - Q2 - T2
Categ. JCR: ENGINEERING, ELECTRICAL & ELECTRONIC rank: 74 / 274 = 0.27 (2022) - Q2 - T1

Factor impacto CITESCORE: 6.4 - Earth and Planetary Sciences (Q1) - Engineering (Q1)

Factor impacto SCIMAGO: 1.284 - Geotechnical Engineering and Engineering Geology (Q1) - Electrical and Electronic Engineering (Q1)

Financiación: info:eu-repo/grantAgreement/ES/AEI-FEDER/PID2019-105660RB-C21
Financiación: info:eu-repo/grantAgreement/ES/DGA/T58-17R
Financiación: info:eu-repo/grantAgreement/ES/MINECO/TIN2016-76635-C2-1-R

Tipo y forma: Article (PostPrint)

Área (Departamento): Área Arquit.Tecnología Comput. (Dpto. Informát.Ingenie.Sistms.)