Prof. Dr. Paul Scheunders
Professor & Remote Sensing Lead • Vision Lab
Biography
Paul Scheunders graduated in physics at the University of Antwerp in 1986. He obtained a PhD in statistical physics at the same university in 1990. From 1989 to 1990, he worked in the Physics Department at the University of Louvain. From 1990 to 1992, he worked in the Nuclear Medicine division of the University Hospital of Louvain. In 1992 he and Prof. Dirk Van Dyck started the Vision Lab at the University of Antwerp, where he is currently a Professor.
Research
My general research interest is in the development of techniques for image processing and machine learning with a strong emphasis on model-based and statistical techniques. My main application area is remote sensing in general and multispectral and hyperspectral image processing and analysis in particular. Recent work shifted towards close-range applications, for which we use our own acquisition equipment: a spectrometer (400-2500 nm), a VNIR hyperspectral camera (400-900 nm) and a SWIR hyperspectral camera (1000-1700 nm).
Some specific research topics are:
- Development of lineair and non-linear spectral unmixing methods
- Material characterization: material composition, water content estimation, biophysical parameter estimation
- Applications related to clay and powder composition estimation
- Applications related to moisture content of soil and construction materials
- Applications related to corrosion monitoring
Teaching
- Mathematical Methods for Physics II; 1-st Bachelor Physics
- Introduction Classic Field Theory; 2-th Bachelor Physics
- Theoretical Physics: Classic Field Theory; 3-th Bachelor Physics
- Artificial Neural Networks; Master Physics
Selected Publications
- Loading selected publications...
PhD Theses Advised
-
Hyperspectral imaging for automated inspection of offshore wind infrastructure
-
Unsupervised machine/deep learning mapping (clustering) for single/multi-source remote sensing data
-
Development of advanced hyperspectral unmixing methods
-
Hyperspectral image mixture analysis using notions of sparsity, nonlinearity and decision fusion
-
Analysis of hyperspectral images for high-throughput plant phenotyping
-
Advances in unmixing of hyperspectral remote sensing imagery
-
Data-driven methods for the analysis of time-resolved mental chronometry fMRI data sets : approaches and techniques based on the Fuzzy Clustering Method and spatial Independent Component Analysis
-
Reconstructie van Transversaal Getrunceerde Cone Beam Projecties in Micro-Tomografie
-
Feature Selection and Classifier Ensembles: A Study on Hyperspectral Remote Sensing Data
-
Unsupervised Pattern Recognition - Dimensionality Reduction and Classification
-
Image Analysis for Material Characterisation