Premium
A fuzzy logic system for band detection in Raman spectroscopy
Author(s) -
PerezPueyo R.,
Soneira M. J.,
RuizMoreno S.
Publication year - 2004
Publication title -
journal of raman spectroscopy
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.748
H-Index - 110
eISSN - 1097-4555
pISSN - 0377-0486
DOI - 10.1002/jrs.1178
Subject(s) - raman spectroscopy , fuzzy logic , wavenumber , a priori and a posteriori , position (finance) , g band , optics , artificial intelligence , algorithm , computer science , physics , philosophy , epistemology , finance , economics
Raman spectroscopy can be used to identify artistic materials on the basis of their characteristic Raman spectra. The acquisition of the wavenumber position of the Raman bands is frequently carried out by visual inspection. Obviously this is a slow, imprecise and non‐automatic process. In this paper, a fuzzy logic system to locate Raman bands in an automated way, avoiding subjective errors, is presented. The proposed fuzzy logic system is based on a set of rules, which give the processing strategy for detecting the wavenumber position of Raman bands. The fuzzy rule‐based system performs the location of Raman bands in a spectrum in the same way as an observer who examines a spectrum visually and tries to localize the Raman bands. That is, he/she compares the possible bands that he has seen with a function that, as known a priori , is representative of a Raman band. Depending on the comparison, he/she establishes that the peak is actually a Raman band. Bearing this in mind, the fuzzy Raman band location system is based on the following statement: when some zone of the spectrum stands out from the nearby background with a characteristic shape, it is established that this zone contains a distinguishable Raman band. In order to know when one zone of the spectrum ‘stands out from the nearby background’, the parabolic approximation of the spectrum in this zone is calculated. The band detection is based on this approximation. The membership functions, which characterize the fuzzy sets at the input and output of the system, and the inference mechanism suitable for the problem are chosen. In the visual process, the knowledge and experience of the observer control the decision‐making process; however, fuzzy logic provides a powerful technique in the decision‐making environment since it incorporates this knowledge. Copyright © 2004 John Wiley & Sons, Ltd.