Dimensionality Reduction Algorithms for Large-Scale ImagesMS53

Technological and theoretical advances in all scientific disciplines ranging from Engineering to Sciences have provided us with numerous large scale datasets to analyze. A fundamental question is how extract low dimensional models that will reflect in an efficient manner the most-important states and dynamics of the system under study. To tackle with such problems this symposium confront with two challenges: that of dimensionality reduction and the problem of data mining. Algorithms and methods that have the potential to facilitate better understanding, predicting and modelling of large-scale data and images with important health, social and economical impact will be discussed.

On the generation of reduced models by Proper Orthogonal Decomposition from experimental image data
Lucia Russo (Consiglio Nazionale delle Ricerche, Istituto di Ricerche sulla Combustione)
Intrinsic Isometric Manifold Learning with Application to Unsupervised Localization from Image Data
Ariel Schwartz (Technion - Israel Institute of Technology)
Construction of low dimensional Functional connectivity networks from fMRI data using manifold learning algorithms
Costantinos Siettos (National Technical University of Athens )
Contrast enhancement operators based on attractors identification in nonlinear dynamical systems
Jacques Demongeot (University of Grenoble Alpes (UGA))
Neural Manifolds: Sparse Dictionary Learning Approaches
Francesco Donnarumma (Institute of Cognitive Sciences and Technologies, Rome)
Application of the Optical Flow Method for the analysis of flame propagation in a transparent internal combustion engine
Simone Lombardi (Università degli Studi del Sannio)
Application of Decomposition Methods to the study of flames via image sequences
Gaetano Continillo (University of Sannio, Benevento)
Salvatore Cuomo (Dept. Mathematics and Applications "Renato Caccioppoli", University of Naples)
Lucia Russo (Consiglio Nazionale delle Ricerche, Istituto di Ricerche sulla Combustione)
Costantinos Siettos (National Technical University of Athens )
artificial intelligence techniques, bayesian methods, computer graphics, image enhancement, image from combustion problems, image representation, inverse problems, machine learning, pattern recognition