Fusion of 3D point clouds and hyperspectral data for the extraction of geometric and radiometric features of trees

Directeur de thèse :     Jocelyn CHANUSSOT

Co-directeur de thèse :     Mauro DALLA MURA

École doctorale : Electronique, electrotechnique, automatique, traitement du signal (EEATS)

Spécialité : Signal, image, parole, télécoms

Structure de rattachement : Autre

Établissement d'origine : Autre

Financement(s) : Bourse attribuée par un organisme

Date d'entrée en thèse : 01/11/2017

Date de soutenance : 17/12/2020

Composition du jury :

Florence Tupin, Professeur, TelecomParisTech, Présidente
Nesrine Chehata, MCF HDR, Bordeaux INP, Rapportrice
Sylvie Durrieu, DR INRAE Montpellier, Rapportrice
Markus Hollaus, MCF HDR, TU Wien, Autriche, Examinateur
Emmanuel Trouvé, Professeur, Université Savoie Mont Blanc, Examinateur
Jocelyn Chanussot, Professeur, Grenoble INP, directeur de thèse
Invités:
Mauro Dalla mura, MCF, Grenoble INP, co-encadrant
Matthieu Monnet, IR INRAE Grenoble, co-encadrant
Jean-Baptiste Barré, IR IGE Grenoble, co-encadrant

Résumé :
The main topic of this PhD is the fusion of 3D point cloud and hyperspectral data for the extraction of geometric and radiometric features of forest trees. We present the integration of remotely sensed data for the analysis of forest areas. In particular, we focus our attention on hyperspectral and LiDAR data that are of primary importance in the study of forest areas. Our attention is also devoted to the use of unsupervised and supervised machine learning techniques for the use of the information contained in such data acquired over forest areas. To summarize, the thesis objectives answer the three scientific challenges: Q1. How data processing methods are applied in each level of data fusion for forest monitoring? Q2. How a crown shape model can improve the segmentation of individual tree crowns? Q3. Which combination of feature sets contribute to characterize the forest tree species composition?