Comparison of spectra processing methods for SERS based quantitative analysis

doi:10.1109/ICCSS.2017.8091399

	Comparison of spectra processing methods for SERS based quantitative analysis
	Wu Y.; Chen L.
	2017-10-31
Source Publication	ICCSS 2017 - 2017 International Conference on Information, Cybernetics, and Computational Social Systems
Pages	130-136
Abstract	Raman spectroscopy, which can provide valuable information about chemical structure, is widely used as an analytical tool for material identification and quantitative chemical/biological analysis. Surface-enhanced Raman spectroscopy (SERS), overcoming the traditional limitation of weak signals, improves the signal intensity significantly. The preprocessing of Raman spectrum is a very useful way to reduce the irrelevant systematic variations in the spectra data while improves the accuracy of further quantitative analysis, which includes de-nosing, background correction, normalization and intensity measurement. In this paper, we first describe several commonly used processing methods, then check which processing procedure can provide the best regression analysis of peak intensity. The value of peak intensity is correspondent to the concentration of analyte. According to our experiments for the real Raman spectra data collected from detection on low concentration of Rhodamine and an unique internal standard, the procedure which combines the morphological weighted penalized least square(mpls) algorithm with maximum peak height measurement method achieves the best performance among all the studied processing methods.
DOI	10.1109/ICCSS.2017.8091399
URL	View the original
Language	英語
全文获取链接	View Full-Text via DOI View Full-Text via Scopus
引用统计
Document Type	Conference paper
专题	University of Macau
Affiliation	Universidade de Macau
推荐引用方式 GB/T 7714	Wu Y.,Chen L.. Comparison of spectra processing methods for SERS based quantitative analysis[C],2017:130-136.