TanDEM-X Forest Mapping Using Convolutional Neural Networks

In this work, we face the problem of forest mapping from TanDEM-X data by means of Convolutional Neural Networks (CNNs). Our study aims to highlight the relevance of domain-related features for the extraction of the information of interest thanks to their joint nonlinear processing through CNN. In p...

Full description

Saved in:
Bibliographic Details
Published in:Remote sensing (Basel, Switzerland) Switzerland), 2019-12, Vol.11 (24), p.2980
Main Authors: Mazza, Antonio, Sica, Francescopaolo, Rizzoli, Paola, Scarpa, Giuseppe
Format: Article
Language:English
Subjects:
Artificial neural networks
Backscattering
Classification
Data integration
Deep learning
Feature extraction
Forests
Incidence angle
Information retrieval
Interferometric synthetic aperture radar
Mapping
Neural networks
Remote sensing
Signal processing
Vegetation
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this work, we face the problem of forest mapping from TanDEM-X data by means of Convolutional Neural Networks (CNNs). Our study aims to highlight the relevance of domain-related features for the extraction of the information of interest thanks to their joint nonlinear processing through CNN. In particular, we focus on the main InSAR features as the backscatter, coherence, and volume decorrelation, as well as the acquisition geometry through the local incidence angle. By using different state-of-the-art CNN architectures, our experiments consistently demonstrate the great potential of deep learning in data fusion for information extraction in the context of synthetic aperture radar signal processing and specifically for the task of forest mapping from TanDEM-X images. We compare three state-of-the-art CNN architectures, such as ResNet, DenseNet, and U-Net, obtaining a large performance gain over the baseline approach for all of them, with the U-Net solution being the most effective one.
ISSN:2072-4292
2072-4292
DOI:10.3390/rs11242980