Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ ENEA Open Archivearrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
ENEA Open Archive
Other ORP type . 2019
Data sources: ENEA Open Archive
addClaim

This Research product is the result of merged Research products in OpenAIRE.

You have already added 0 works in your ORCID record related to the merged Research product.
addClaim

This Research product is the result of merged Research products in OpenAIRE.

You have already added 0 works in your ORCID record related to the merged Research product.

Evaluation of artificial neural networks performances on spectral classification tasks

Authors: Angelini, F.; Di Frischia, S.;

Evaluation of artificial neural networks performances on spectral classification tasks

Abstract

Questo documento presenta uno studio preliminare riguardo la valutazione delle performance di una rete neurale artificiale (ANN) in uno scenario di classificazione spettrale. Differenti architetture di reti neurali sono state testate al fine di risolvere problemi di identificazione spettrale, iniziando da un semplice insieme di input, per poi raffinare il problema un passo alla volta. Dato che l’obiettivo in futuro è quello di definire un sistema robusto ed efficiente per la classificazione di spettri Raman in condizioni non di laboratorio, l’insieme di spettri di input è stato costruito sul modello di spettri Raman, con segnali rumorosi contenenti picchi gaussiani stretti. Il comportamento delle reti neurali è stato valutato variando le principali proprietà di input e analizzando gli indicatori chiave di performance delle reti neurali. Tutti i test contenuti in questo documento sono stati sviluppati e generati con il software MATLAB®. Dopo una descrizione generale delle reti neurali e delle loro proprietà, il documento analizza l’identificazione del problema, la definizione di rumore e alcuni concetti nel campo dell’inferenza statistica. Quindi, vengono illustrati una serie di test con i relativi grafici di performance, in cui sono stati usati sia spettri artificiali che spettri reali generati in laboratorio con una lampada a vapori di mercurio. Alla fine del documento sono state prodotte una serie di conclusioni, con ulteriori analisi e raffinamenti del problema da eseguire in futuro.

This report is a preliminary study about the evaluation of the Artificial Neural Network (ANN) performances in a spectral classification scenario. Several different ANN architectures has been tested in order to solve signal identification tasks, starting from a simple input set and refining the task step to step. Since the aim is to define in the future a robust method to classify Raman spectra in non-laboratory conditions, the input set has been built as a Raman-like spectra set, with noisy signals containing narrow Gaussian peaks. The behavior of the neural networks has been evaluated varying the main input properties and analyzing the standard ANN key performance indicators. All the tests in this paper have been developed and generated with MATLAB®. After a description of the artificial neural networks and their features, the report analyzes the task identification, the noise definition and some concepts in the statistic theory field. Then, a series of tests with their performance plots are illustrated, both with artificial spectra and real laboratory spectra generated with a mercury-vapor lamp. In the end, the report produces a list of conclusions with some further analysis and task refinements to do.

Country
Italy
Keywords

Signal processing, Algorithm, Artificial intelligence, Raman;Artificial intelligence;Spectroscopy;Signal processing;Algorithms;Spectral classification;Pattern recognition;Neural networks, Pattern recognition, Spectral classification, Raman, Spectroscopy, Algorithms, Neural networks

  • BIP!
    Impact byBIP!
    citations
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
  • citations
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
    Powered byBIP!BIP!
Powered by OpenAIRE graph
Found an issue? Give us feedback
citations
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Related to Research communities
moresidebar

Do the share buttons not appear? Please make sure, any blocking addon is disabled, and then reload the page.