Francescopaolo Sica

Dr. Francescopaolo Sica

German Aerospace Center (DLR) - Microwaves and Radar Institute
orcid

Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 September 2020 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Open access
Francescopaolo SicaFrancescopaolo Sica
Francescopaolo Sica
GGGiorgia Gobbi
Giorgia Gobbi
PRPaola Rizzoli
Paola Rizzoli
LBLorenzo Bruzzone
Lorenzo Bruzzone
Nowadays, deep learning (DL) finds application in a large number of scientific fields, among which the estimation and the enhancement of signals disrupted by the noise of different natures. In this article, we address the problem of the estimation of the interferometric parameters from synthetic aperture radar (SAR) data. In particular, we combine convolutional neural networks together with the concept of residual learning to define a novel architecture, named ϕ-Net, for the joint estimation of the interferometric phase and coherence. ϕ-Net is trained using synthetic data obtained by an innovative strategy based on the theoretical modeling of the physics behind the SAR acquisition principle. This strategy allows the network to generalize the estimation problem with respect to: 1) different noise levels; 2) the nature of the imaged target on the ground; and 3) the acquisition geometry. We then analyze the ϕ-Net performance on an independent data set of synthesized interferometric data, as well as on real InSAR data from the TanDEM-X and Sentinel-1 missions. The proposed architecture provides better results with respect to state-of-the-art InSAR algorithms on both synthetic and real test data. Finally, we perform an application-oriented study on the retrieval of the topographic information, which shows that ϕ-Net is a strong candidate for the generation of high-quality digital elevation models at a resolution close to the one of the original single-look complex data.

21 October 2020 in IEEE Geoscience and Remote Sensing Letters
Verified
Journal Article
Open access
Francescopaolo SicaFrancescopaolo Sica
Francescopaolo Sica
FCFrancesco Calvanese
Francesco Calvanese
Giuseppe ScarpaGiuseppe Scarpa
Giuseppe Scarpa
PRPaola Rizzoli
Paola Rizzoli
Phase unwrapping (PU) is among the most critical tasks in synthetic aperture radar (SAR) interferometry (InSAR). Due to the presence of noise, the interferogram usually presents phase inconsistencies, also called residues, which imply a nonunivocal solution. This work investigates the PU problem from a semantic segmentation perspective by exploiting convolutional neural network (CNN) models. In particular, by exploiting a popular deep-learning architecture, we introduce the interferometric coherence as an input feature and analyze the performance increase against classical methods. For the network training, we generate a variegated data set by introducing a controlled number of phase residues, and considering both synthetic and real InSAR data. Eventually, we compare the proposed method to state-of-the-art algorithms on synthetic and real InSAR data taken from the TanDEM-X mission, obtaining encouraging results.

9 August 2023 in IEEE Geoscience and Remote Sensing Magazine
Verified
Journal Article
Closed access
MSMichael Schmitt
Michael Schmitt
S.Ali AhmadiS.Ali Ahmadi
S.Ali Ahmadi
Yonghao XuYonghao Xu
Yonghao Xu
Gulsen TaskinGulsen Taskin
Gulsen Taskin
UVUjjwal Verma
Ujjwal Verma
RHRonny Hänsch
Ronny Hänsch
Carefully curated and annotated datasets are the foundation of machine learning (ML), with particularly data-hungry deep neural networks forming the core of what is often called artificial intelligence ( AI ). Due to the massive success of deep learning (DL) applied to Earth observation (EO) problems, the focus of the community has been largely on the development of evermore sophisticated deep neural network architectures and training strategies. For that purpose, numerous task-specific datasets have been created that were largely ignored by previously published review articles on AI for EO. With this article, we want to change the perspective and put ML datasets dedicated to EO data and applications into the spotlight. Based on a review of historical developments, currently available resources are described and a perspective for future developments is formed. We hope to contribute to an understanding that the nature of our data is what distinguishes the EO community from many other communities that apply DL techniques to image data, and that a detailed understanding of EO data peculiarities is among the core competencies of our discipline.

12 December 2019 in Remote Sensing
Verified
Journal Article
Open access
PDF file
Antonio MazzaAntonio Mazza
Antonio Mazza
Francescopaolo SicaFrancescopaolo Sica
Francescopaolo Sica
PRPaola Rizzoli
Paola Rizzoli
Giuseppe ScarpaGiuseppe Scarpa
Giuseppe Scarpa
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.

6 March 2020 in Remote Sensing
Verified
Journal Article
Open access
PDF file
Andrea PulellaAndrea Pulella
Andrea Pulella
RSRodrigo Aragão Santos
Rodrigo Aragão Santos
Francescopaolo SicaFrancescopaolo Sica
Francescopaolo Sica
PPPhilipp Posovszky
Philipp Posovszky
PRPaola Rizzoli
Paola Rizzoli
This paper reports recent advancements in the field of Synthetic Aperture Radar (SAR) for forest mapping by using interferometric short-time-series. In particular, we first present how the interferometric capabilities of the Sentinel-1 satellites constellation can be exploited for the monthly mapping of the Amazon rainforest. Indeed, the evolution in time of the interferometric coherence can be properly modeled as an exponential decay and the retrieved interferometric parameters can be used, together with the backscatter, as input features to the machine learning Random Forests classifier. Furthermore, we present an analysis on the benefits of the use of textural information, derived from Sentinel-1 backscatter, in order to enhance the classification accuracy. These textures are computed through the Sum And Difference Histograms methodology and the final classification accuracy, resulting by adding them to the aforementioned features, is a thematic map that exceeds an overall agreement of 85 % , when validated using the optical external reference Finer Resolution Observation and Monitoring of Global Land Cover (FROM-GLC) map. The experiments presented in the final part of the paper are enriched with a further analysis and discussion on the selected scenes using updated multispectral Sentinel-2 acquisitions.

14 December 2021 in Remote Sensing
Verified
Journal Article
Open access
PDF file
JBJosé-Luis Bueso-Bello
José-Luis Bueso-Bello
Michele MartoneMichele Martone
Michele Martone
CGCarolina Gonzalez
Carolina Gonzalez
Francescopaolo SicaFrancescopaolo Sica
Francescopaolo Sica
PVPaolo Valdo
Paolo Valdo
PRPaola Rizzoli
Paola Rizzoli
The interferometric synthetic aperture radar (InSAR) data set, acquired by the TanDEM-X (TerraSAR-X add-on for Digital Elevation Measurement) mission (TDM), represents a unique data source to derive geo-information products at a global scale. The complete Earth’s landmasses have been surveyed at least twice during the mission bistatic operation, which started at the end of 2010. Examples of the delivered global products are the TanDEM-X digital elevation model (DEM) (at a final independent posting of 12 m × 12 m) or the TanDEM-X global Forest/Non-Forest (FNF) map. The need for a reliable water product from TanDEM-X data was dictated by the limited accuracy and difficulty of use of the TDX Water Indication Mask (WAM), delivered as by-product of the global DEM, which jeopardizes its use for scientific applications, as well. Similarly as it has been done for the generation of the FNF map; in this work, we utilize the global data set of TanDEM-X quicklook images at 50 m × 50 m resolution, acquired between 2011 and 2016, to derive a new global water body layer (WBL), covering a range from −60∘ to +90∘ latitudes. The bistatic interferometric coherence is used as the primary input feature for performing water detection. We classify water surfaces in single TanDEM-X images, by considering the system’s geometric configuration and exploiting a watershed-based segmentation algorithm. Subsequently, single overlapping acquisitions are mosaicked together in a two-step logically weighting process to derive the global TDM WBL product, which comprises a binary averaged water/non-water layer as well as a permanent/temporary water indication layer. The accuracy of the new TDM WBL has been assessed over Europe, through a comparison with the Copernicus water and wetness layer, provided by the European Space Agency (ESA), at a 20 m × 20 m resolution. The F-score ranges from 83%, when considering all geocells (of 1∘ latitudes × 1∘ longitudes) over Europe, up to 93%, when considering only the geocells with a water content higher than 1%. At global scale, the quality of the product has been evaluated, by intercomparison, with other existing global water maps, resulting in an overall agreement that often exceeds 85% (F-score) when the content in the geocell is higher than 1%. The global TDM WBL presented in this study will be made available to the scientific community for free download and usage.