Learning Sentinel-2 Spectral Dynamics for Long-run Predictions using Residual Neural Networks - Archive ouverte HAL Access content directly
Conference Papers Year :

Learning Sentinel-2 Spectral Dynamics for Long-run Predictions using Residual Neural Networks

(1) , (1) , (2, 3)
1
2
3

Abstract

Making the most of multispectral image time-series is a promising but still relatively under-explored research direction because of the complexity of jointly analyzing spatial, spectral and temporal information. Capturing and characterizing temporal dynamics is one of the important and challenging issues. Our new method paves the way to capture real data dynamics and should eventually benefit applications like unmixing or classification. Dealing with time-series dynamics classically requires the knowledge of a dynamical model and an observation model. The former may be incorrect or computationally hard to handle, thus motivating data-driven strategies aiming at learning dynamics directly from data. In this paper, we adapt neural network architectures to learn periodic dynamics of both simulated and real multispectral time-series. We emphasize the necessity of choosing the right state variable to capture periodic dynamics and show that our models can reproduce the average seasonal dynamics of vegetation using only one year of training data.
Fichier principal
Vignette du fichier
2011.08746.pdf (1.24 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03265010 , version 1 (18-06-2021)

Identifiers

Cite

Joaquim Estopinan, Guillaume Tochon, Lucas Drumetz. Learning Sentinel-2 Spectral Dynamics for Long-run Predictions using Residual Neural Networks. EUSIPCO 2021: 29th European Signal Processing Conference, Aug 2021, Dublin (virtual), Ireland. ⟨10.23919/EUSIPCO54536.2021.9616304⟩. ⟨hal-03265010⟩
33 View
40 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More