An experimental analysis of the use of different neural models for the supervised classification of multisensor remote-sensing data is presented. Three types of neural classifiers are considered: the Multilayer Perceptron, a kind of Structured Neural Network, proposed by the authors, that allows the interpretation of the network operation, and a Probabilistic Neural Network. Furthermore, the k-nearest neighbour statistical classifier is also considered in order to evaluate the validity of the aforementioned neural networks, as compared with that of classical statistical methods. The results provided by the above classifiers are compared.
An experimental comparison of neural and statistical non-parametric algorithms for supervised classification of remote-sensing images RID A-2076-2012
ROLI, FABIO
1996-01-01
Abstract
An experimental analysis of the use of different neural models for the supervised classification of multisensor remote-sensing data is presented. Three types of neural classifiers are considered: the Multilayer Perceptron, a kind of Structured Neural Network, proposed by the authors, that allows the interpretation of the network operation, and a Probabilistic Neural Network. Furthermore, the k-nearest neighbour statistical classifier is also considered in order to evaluate the validity of the aforementioned neural networks, as compared with that of classical statistical methods. The results provided by the above classifiers are compared.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
