Premium
Application of the artificial neural network approach to the recognition of specific patterns in Auger electron spectroscopy
Author(s) -
Souza M. N.,
Gatts C.,
Figueira M. A.
Publication year - 1993
Publication title -
surface and interface analysis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.52
H-Index - 90
eISSN - 1096-9918
pISSN - 0142-2421
DOI - 10.1002/sia.740201303
Subject(s) - auger electron spectroscopy , artificial neural network , electron , spectroscopy , electron spectroscopy , materials science , pattern recognition (psychology) , atomic physics , analytical chemistry (journal) , computer science , artificial intelligence , physics , chemistry , nuclear physics , astronomy , environmental chemistry
The artificial neural network (ANN) approach was applied to the identification of Auger electron spectral patterns. The ANN structure employed was the counter‐propagation architecture with an unsupervised learning algorithm. For training such a network, it is only necessary to provide a data set with samples of the patterns to be recognized, and the network itself will extract the relevant statistical information to organize similar patterns into specific classes. We used a training data set of five different Auger spectra (Fe, Au, Si, Sn, Cu) to which a random fluctuation of up to 5% of the highest peak was added. To the test set, however, the added fluctuation was up to 50% and we observed that the network was able to identify precisely and test spectrum after only a few training sessions. The ANN synapses can be interpreted as the average spectra of the training set for each specific class, tending to zero fluctuation spectra as the number of training samples becomes large. The results obtained show that even by using an extremely simple ANN structure the classification of single‐element Auger spectra was made easy also in the case of extremely noisy spectra.