A5002883250- Synthetic Aperture Radar (SAR) Applications and Techniques
- Remote Sensing and LiDAR Applications
- Landslides and related hazards
- Remote Sensing in Agriculture
- Forest ecology and management
- Cryospheric studies and observations
- Advanced SAR Imaging Techniques
- Ecosystem dynamics and resilience
- Soil Moisture and Remote Sensing
- Marine and environmental studies
- Soil erosion and sediment transport
University of Trento
2023-2024
Deutsches Zentrum für Luft- und Raumfahrt e. V. (DLR)
2022-2024
The bistatic interferometric coherence is affected by different sources of error, among which volume decorrelation, quantifies the amount noise caused scattering mechanisms. This represents a key quantity not only for performance assessment synthetic aperture radar (SAR) products, but also large variety scientific applications, ranging from land cover classification to physical parameters estimation, such as ice structure, forest height, and biomass retrieval. magnitude decorrelation can be...
Operational canopy height mapping at high resolution remains a challenging task country-level. Most of the existing state-of-the-art inversion methods propose physically-based schemes which are specifically tuned for local scales. Only few approaches in literature have attempted to produce country or global scale estimates, mostly by means data-driven and multi-spectral data sources. In this paper, we robust deep learning approach that exploits single-pass interferometric TanDEM-X generate...
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 Deep Learning (DL) regression from TanDEM-X bistatic interferometric (InSAR) data. We propose novel fully convolutional neural network (CNN) framework, trained in supervised manner using reference CHM measurements derived LiDAR LVIS airborne sensor NASA. The were acquired during joint NASA-ESA 2016 AfriSAR campaign...
The TanDEM-X synthetic aperture radar (SAR) system allows for the recording of bistatic interferometric SAR (InSAR) acquisitions, which provide additional information to common amplitude images acquired by monostatic systems. More concretely, volume decorrelation factor, can be derived from coherence, is a reliable indicator presence vegetation and it was used as main input feature generation global forest/non-forest map, means clustering algorithm. In this work, we investigate capabilities...
Large-scale and up-to-date canopy height model (CHM) estimates are key to forest resources assessment disturbance analysis. In this work we present an investigation of the potential Deep Learning (DL) for regression from TanDEM-X bistatic InSAR data. We propose a novel fully convolutional neural network (CNN) framework, trained tested on four tropical sites in Gabon, Africa, together with series experiments assessing impact different input features specific focus InSAR. The obtained results...
The estimation of forest parameters, such as canopy height model (CHM) and above ground biomass (AGB), is ut-most importance for monitoring, carbon-cycle modelling, disturbance analysis, resource inventorying natural disaster prevention. In this work, we profit from the most recent advancements in deep learning research to propose a convolutional neural network (CNN) architecture frequent parameter at large scale. Our technique consists fully convolutional, multi-modal framework, which works...
The TanDEM-X Forest/Non-Forest map, derived from the volume decorrelation factor using a supervised fuzzy clustering algorithm, represents baseline approach for forest mapping with data at global scale. Deep learning (DL) methods have been demonstrated to be also suitable forests large scale interferometric data. In this work, we investigate capabilities of U-Net-like architecture 6 m resolution. With such high-resolution data, aim improving accuracy and able detect degradation over Amazon...
In this work, we investigate the potential of SAR interferometry (InSAR) for mapping forests worldwide and retrieve important biophysical parameters, such as land cover canopy height. We compare single-pass (bistatic) versus repeat-pass InSAR, discussing their main peculiarities limitations. particular, concentrate on analysis interferometric coherence relationship between volume temporal decorrelation with respect to forest parameters estimation. present work done at DLR high spatial...
<p>Covering about 30 percent of the Earth’s surface, forests are paramount importance for ecosystem. They act as effective carbon sinks, reducing concentration greenhouse gas in atmosphere, and help mitigating climate change effects. This delicate ecosystem is currently threatened degraded by anthropogenic activities natural hazards, such deforestation, agricultural activities, farming, fires, floods, winds, soil erosion. In an era dramatic changes ecosystems,...
The TanDEM-X Forest/Non-Forest Map, derived from the volume decorrelation factor using a supervised fuzzy clustering algorithm, represents baseline approach for forest mapping with data at large/global-scale. Deep learning methods have been demonstrated to be also suitable forests interferometric data, e.g. by utilizing U-Net convolutional neural network (CNN) on full-resolution images. In this work, we investigate capabilities of U-Net-like architecture and water large scale. An ad-hoc...