DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

A Deep Learning Framework for the Estimation of Forest Height From Bistatic TanDEM-X Data

Carcereri, Daniel and Rizzoli, Paola and Ienco, Dino and Bruzzone, Lorenzo (2023) A Deep Learning Framework for the Estimation of Forest Height From Bistatic TanDEM-X Data. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. IEEE - Institute of Electrical and Electronics Engineers. doi: 10.1109/JSTARS.2023.3310209. ISSN 1939-1404.

[img] PDF - Published version


Up-to-date canopy height model (CHM) estimates are of key importance for forest resources monitoring and disturbance analysis. In this work we present a study on the potential of Deep Learning (DL) for the regression of forest height from TanDEM-X bistatic interferometric (InSAR) data. We propose a novel fully convolutional neural network (CNN) framework, trained in a supervised manner using reference CHM measurements derived from the LiDAR LVIS airborne sensor from NASA. The reference measurements were acquired during the joint NASA-ESA 2016 AfriSAR campaign over five sites in Gabon, Africa, characterized by the presence of different kinds of vegetation, spanning from tropical primary forests to mangroves. Together with the DL architecture and training strategy, we present a series of experiments to assess the impact of different input features on the network estimation accuracy (in particular of bistatic InSAR-related ones). When tested on all considered sites, the proposed DL model achieves an overall performance of 1.46m mean error, 4.2m mean absolute error and 15.06% mean absolute percentage error. Furthermore, we perform a spatial transfer analysis aimed at deriving preliminary insights on the generalization capability of the network when trained and tested on data sets acquired over different locations, combining different kinds of tropical vegetation. The obtained results are promising and already in line with state-of-the-art methods based on both physical-based modelling and data-driven approaches, with the remarkable advantage of requiring only one single TanDEM-X acquisition at inference time.

Item URL in elib:https://elib.dlr.de/197676/
Document Type:Article
Title:A Deep Learning Framework for the Estimation of Forest Height From Bistatic TanDEM-X Data
AuthorsInstitution or Email of AuthorsAuthor's ORCID iDORCID Put Code
Carcereri, DanielUNSPECIFIEDhttps://orcid.org/0000-0002-3956-1409UNSPECIFIED
Rizzoli, PaolaUNSPECIFIEDhttps://orcid.org/0000-0001-9118-2732UNSPECIFIED
Ienco, DinoUNSPECIFIEDhttps://orcid.org/0000-0002-8736-3132UNSPECIFIED
Bruzzone, LorenzoUNSPECIFIEDhttps://orcid.org/0000-0002-6036-459XUNSPECIFIED
Date:30 August 2023
Journal or Publication Title:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Refereed publication:Yes
Open Access:Yes
Gold Open Access:Yes
In ISI Web of Science:Yes
Publisher:IEEE - Institute of Electrical and Electronics Engineers
Keywords:bistatic coherence, convolutional neural network, deep learning, forest height, synthetic aperture radar, synthetic aperture radar interferometry, TanDEM-X
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Space
HGF - Program Themes:Earth Observation
DLR - Research area:Raumfahrt
DLR - Program:R EO - Earth Observation
DLR - Research theme (Project):R - AI4SAR
Location: Oberpfaffenhofen
Institutes and Institutions:Microwaves and Radar Institute
Microwaves and Radar Institute > Spaceborne SAR Systems
Deposited By: Carcereri, Daniel
Deposited On:04 Oct 2023 15:56
Last Modified:22 Feb 2024 09:25

Repository Staff Only: item control page

Help & Contact
electronic library is running on EPrints 3.3.12
Website and database design: Copyright © German Aerospace Center (DLR). All rights reserved.