Learning Endmember Dynamics in Multitemporal Hyperspectral Data Using A State-Space Model Formulation - IMT Atlantique Access content directly
Conference Papers Year : 2020

Learning Endmember Dynamics in Multitemporal Hyperspectral Data Using A State-Space Model Formulation

Abstract

Hyperspectral image unmixing is an inverse problem aiming at recovering the spectral signatures of pure materials of interest (called endmembers) and estimating their proportions (called abundances) in every pixel of the image. However, in spite of a tremendous ap-plicative potential and the avent of new satellite sensors with high temporal resolution, multitemporal hyperspectral unmixing is still a relatively underexplored research avenue in the community, compared to standard image unmixing. In this paper, we propose a new framework for multitemporal unmixing and endmember extraction based on a state-space model, and present a proof of concept on simulated data to show how this representation can be used to inform multitemporal unmixing with external prior knowledge, or on the contrary to learn the dynamics of the quantities involved from data using neural network architectures adapted to the identification of dynamical systems.
Fichier principal
Vignette du fichier
drumetz_endmember_dynamics_2019.pdf (534.63 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-02490607 , version 1 (25-02-2020)

Identifiers

Cite

Lucas Drumetz, Mauro Dalla Mura, Guillaume Tochon, Ronan Fablet. Learning Endmember Dynamics in Multitemporal Hyperspectral Data Using A State-Space Model Formulation. ICASSP 2020 - IEEE International Conference on Acoustics, Speech and Signal Processing, IEEE, May 2020, Barcelone (virtual), Spain. ⟨10.1109/ICASSP40776.2020.9053787⟩. ⟨hal-02490607⟩
114 View
117 Download

Altmetric

Share

Gmail Facebook X LinkedIn More