A METHOD FOR EXTRACTING INFORMATION OF INTEREST FROM MULTI-DIMENSIONAL, MULTI -PARAMETRIC AND/OR MULTI -TEMPORAL DATASETS

摘要

Method for the extraction of information of interest to multi-dimensional, multi -parametric and / or multi temporal datasets related to a same object under observation by means of data fusion in which a plurality of different data sets are provided concerning a single object and with the data related to various parameters and / or at different time acquisition instants of said parameters; the said data set are subjected to a first processing step by means of principal component analysis (PCA so-called) which are generated by an identical number of datasets with transformed data and represented by a combination of "feature"; each of said datasets is combined in a non linear way with the corresponding transformed' data set to obtain a certain predetermined number of datasets combination of parameters by means of weighing; weighting parameters which are determined in an empirical experimental way by means of which the training datasets which are used to determine the values of the non-linear weighting parameters that maximize the value of the new features associated with the data of interest, as compared to those of other data.