A neural network approach for micro-PIXE data

Authors
Rokita, E Mutsaers, PHA Quaedackers, JA Tabor, Z de Voigt, MJA
Citation
E. Rokita et al., A neural network approach for micro-PIXE data, NUCL INST B, 158(1-4), 1999, pp. 159-164
Citations number
12
Categorie Soggetti
Spectroscopy /Instrumentation/Analytical Sciences","Instrumentation & Measurement
Journal title
NUCLEAR INSTRUMENTS & METHODS IN PHYSICS RESEARCH SECTION B-BEAM INTERACTIONS WITH MATERIALS AND ATOMS
ISSN journal
0168583X → ACNP
Volume
158
Issue
1-4
Year of publication
1999
Pages
159 - 164
Database
ISI
SICI code
0168-583X(199909)158:1-4<159:ANNAFM>2.0.ZU;2-W
Abstract
The application of the artificial neural network for the processing of one- dimensional micro-PIXE data is described. The network architecture, selecti on of the transfer function as well as the training and verification operat ions are described in detail. The performed reconstructions confirm that th e neural network may be used for improvement of the resolution and for proc essing of low statistics data. The limitations of the neural network applic ation for two-dimensional images are discussed. (C) 1999 Elsevier Science B .V. All rights reserved.