Degree in Physics cum laude at the University of Napoli in 1981.
From 1981 to 1984 employee at the “Azienda per il Risparmio Energetico” in the R&D sector; from 1984 to 1986 Professor at High School in Electronics and Science. From 1986 Researcher at the Italian National Research Council. From 2016 Director of research at the Italian National Research Council.
Manager of the Istituto per le Applicazioni del Calcolo ‘Mauro Picone’ CNR, Napoli branch from 2002 to 2003 and from 2006 to 2014.
Elected by the CNR researchers and technologists as a member of the CNR Management Board, but not selected by the Italian Minist ry of University and Research (2012).
PI of 8 projects on the development of statistical methodologies in the framework of bilateral cooperation agreements between CNR and corresponding French, Rumanian, Chinese and Israel Agencies (1996-2005). PI of the project "Il sistema scolastico come sistema complesso: qualità delle rilevazioni e modelli di interpretazione dei risultati", INVALSI, 2006-2008
Responsible of Units in the following projects: “Proba-V Clouds Detection Round Robin Protocol” (European Space Agency, 2016-2017); “ACTIVITI - Attrattori Culturali e
Tecnologie Informatiche per la Valorizzazione Interattiva e il Turismo Innovativo” (District of Campania, 2013-2015); “BERSAGLI: sonde e segnali in terapia diagnostica”
(District of Campania, 2013-2015); “Sviluppo di metodologie per l’estrazione e l’integrazione delle informazioni diagnostiche finalizzate a definire percorsi clinici terapeutici
personalizzati in patologie ad elevato impatto sociale” in the framework of the programme MERIT (Italian Ministry of Education and Research, 2011-2014); “SAP4PRISMA -
Sviluppo di Algoritmi e Prodotti per applicazioni in agricoltura ed il monitoraggio del territorio a supporto della missione PRISMA” (Italian Space Agency, 2011-2015).
Responsible of 15 fellowships at the Italian National Research Council.
Tutor of 15 theses at the University of Napoli.
Referee of numerous international journals.
Member of the Organizing Committee of the following International Conferences: MASCOT NUM 2011, Villard-de-Lans, France, 2011; First International Conference on VLSI,
Communication, Computation and Security (ICVCCS’10), Karunya Nagar (India), 2010; BIOCOMP 2005 – Diffusion processes in Neurobiology and Subcellular Biology, Vietri
sul Mare, Italy, 2005; Grenoble 1994-2003 – Wavelets & Statistics, Watering the seed, Villard de Lans, France, 2003; BIOCOMP 2002 - Topics in Biomathematics and related
computational problems at the be-ginning of the Third Millennium, Vietri sul Mare, Italy, 2002
Reviewer of projects for the University of Hong-Kong.
Active in divulgating science at specific events, particu larly “Futuro Remoto”, yearly organized by the Science Center Città della Scienza, Napoli, and “Giochi senza barriere”,
yearly organized by the no-profit organization Tutti a scuola, Napoli.
Author of about 120 peer-reviewed publications on International jou rnals.
Sceintific activity has been devoted to the development of mathematical a nd statistical methods and to their application to several fields:
- Application of statistical methods to the analysis of climatic data. Analysis of solar radiation from the climatological and statistical point of view; development of a
simplified method for estimating solar spectrum.
- Energy systems. Development of a model that s imulates a generic photochemical converter of solar energy; development of a model that simulates the thermal behaviour
of a building. Optimization of districting heating plants. Analysis of the Italian Energy System and assessment of the renewable energy sources and energy saving measures
in different fuel scenarios. The methodology is largely based on linear programming tools.
- Numerical methods for partial differential equations with applications. Simulation of air m otion inside cylinders of engines aimed at reducing the cost of development of
engines and abating pollutant emissions. Development of a mixed implicit-semilagrangian methodology based on finite volumes and conformal mapping from a timevarying
domain to a fixed one. Development of a fractional step mixed explicit/semi-implicit methodology. Development of a method for approximating equations of a
shallow water model applied to atmospheric circulation.
- Numerical and statistical methods for atmospheric phy sics and remote sensing. Development of parametric and nonparametric methods for retrieving size distribution of
aerosols in the atmosphere starting from measures of solar radiation at ground. In particular a regularization methodology has been developed where the penalization term
is given by entropy for which convergence has been proved. Methods have been developed for retrieving atmospheric profiles of temperature, humidity and concentration of
selected gas constituents from interferograms produced by specific infrared radiometers (High-resolution Interferometry Sounders). Initially a technique has been developed
for estimating spectrum for interferograms. From the mathematical point of view the problem is governed by a cosine Fourier transform, that is solved in a framework of
inverse problems and therefore uses regularization tools. The regularization parameter is chosen by Generalized Cross Validation (GCV). The retrieval of geophysical
parameters from spectra is governed by a first kind Fredholm integral equation that has been solved by Generalized Singular Value Decomposition (GSVD) and Truncated
Generalized Singular Value Decomposition (TGSVD) tools. An analytical model has been developed of the Radiative Transfer Equation (RTE) in the atmosphere that also yields
an analytical calculus of the Jacobians of the equation. The model is based on a specific parameterization of the atmospheric optical depth at very high spectral resolution
with respect to temperature and its compression by wavelet packets. Its implementation is able to dramatically abat the computational time needed for computing radiance,
which is essential for retrieving geophysical parameters in real time. From the methodological point of view a semiiterative Landweber method has been developed for
solving the RTE with a suitable stop criterion. The researches have been developed in the framework of international projects on sensors actually flying onboard satellites
(IASI from EUMETSAT and IMG from the Japanese NASDA), for which they have been used to assess the performance of the retrieved geophysical products. A method has
been developed based on wavelets and on wavelet packets able to compress images taken from AVHRR sensor onboard satellites.
- Statistical methods for image processing with applications. Statistical methods have been developed that are specific for images a nd applied to different fields in remote
sensing and medicine, aimed at retrieving important features from the images. In all cases the statistical problem involved is classification which is solved by a class of
methods based on Discriminant Analysis. In particular the problem of detecting clouds from the multi- or hyper-spectral images has been considered. It has been applied to
several sensors: AVHRR, SEVIRI, MODIS, PRISMA, Proba-V. It has been chosen by EUMETSAT as the reference product for the next generation METEOSAT satellites
(METEOSAT Third Generation, MTG). The problem of estimating the surface typology (e.g., bare soil, water, urban, agricultural) from images taken from sensors onboard
satellite (LANDSAT) or airborne (MIVIS sensor) has also been considered. In the medical field the problem of detecting pathological conditions in Magnetic Resonance (MR)
Images has been considered.
- Statistical methodologies. T he method of regularization by Entropy functional has been pioneering introduced for the solution of First Kind Fredholm Integral equations.
The corresponding paper U. Amato, W.P. Hughes: Maximum Entropy Regularization of Fredholm Integral equations of the First Kind, Inverse Problems 7, 793-808 (1991) has
been selected by the journal Inverse Problems as representative of year 1991 for the celebrations of the 25th year of the journal. The problem of nonparametric regression
has been considered: a method based on the Minimum Noise Fraction (MNF) has been developed specific for a problem of abundancy estimation in remote sensing; the
method of wavelet regularization has been developed having optimal theoretical properties among linear methods. A method based on wavelets has also been developed
able to deal with nonequispaced and nondyadic data. Specific regression methods have been developed for the case of multivariate time series, relying on splines and
wavelets for smooth and nonsmooth components, with particular penalization functionals that preserve sparseness of the solution by groups. In this respect the solutions
also aim at reducing the dimension of the problems mitigating the curse of dimensionality. A specific method for the nonparametric estimate of the density function has also
been considered endowed with a Factor Analysis model that can also applied to a classification problem. More methodologies have been considered for the classification
problem based on the Discriminant Analysis endowed with Independent Component Analysis (ICDA), also exploiting the spatial correlation of images (Localized Discriminant
Analysis); for the 2-class problem a method has been developed (Cumulative Discriminant Analysis, CDA) that allows to give solutions having the same Type-I and Type-II
error. This method has been introduced for a competition by ESA on cloud detection from images taken by multispectral sensors onboard satellites by state-of-art
methodologies, ranking the second.