PhD, INRIA Sophia Antipolis
Keywords: Data Reduction , Support Vector Machines, Classification, Statistical Parametric Mapping
Phone: (33)4-97-15-53-68 (inria) / (33)4-92-94-27-77 (I3S)
Postal address: INRIA Sophia Antipolis , 2004, route des Lucioles, 06902 Sophia Antipolis Cedex, France / I3S , 2000, route des Lucioles, 06902 Sophia Antipolis Cedex, France
A hyper-spectral image consists of the same scene pictured at different wavelengths. Such an image enables to analyze an object or a scene by both, its spatial and spectral properties. The aim of my PhD, in collaboration with Galderma (pharmaceutical company specializing in dermatology, whose shareholders are L’Oréal and Nestlé ) is to characterize skin lesions with multi and hyper-spectral images. For this, we are interested in spectral analysis and data reduction tools such as principal component analysis (PCA), independent component analysis (ICA) or projection pursuit, as well as classification methods like support vector machines (SVMs) for example.
My PhD is the following of my Master internship (from February to September 2009).
Sylvain Prigent received the Master Engineering degree from the Department of Electrical Engineering at ENSIEG, Grenoble, France in 2009, and the MSc in Signal and Image processing at the Grenoble-INP, France in 2009. He is currently pursuing a Ph.D degree at INRIA in collaboration with Galderma R&D.
His research topic is multispectral image analysis for the skin lesion severity. His work induced 4 patents by Galderma R&D and INRIA (2 in 2009 and 2 in 2010).
S. Prigent and D. Zugaj and X. Descombes and P. Martel and J. Zerubia. Estimation of an optimal spectral band combination to evaluate skin disease treatment efficacy using multi-spectral images. In Proc. IEEE International Conference on Image Processing (ICIP), Brussels, Belgium, September 2011.