Please use this identifier to cite or link to this item: http://repositorio.unicamp.br/jspui/handle/REPOSIP/32241
Type: Artigo
Title: Determinação de umidade em café cru usando espectroscopia NIR e regressão multivariada
Title Alternative: Determination of moisture in raw coffee by near infra-red reflectance spectroscopy and multivariate regression
Author: Morgano, Marcelo Antonio
Faria, Cristiano Gomes
Ferrão, Marco Flores
Bragagnolo, Neura
Ferreira, Márcia Miguel de Castro
Abstract: Near infra-red reflectance (NIR) spectroscopy was used to measure the moisture content in raw coffee. Different models using partial least squares (PLS) with data pre-processing were used. Regression models were built with 157 spectra of the samples of raw coffee collected using a near infrared spectrometer with an accessory of diffuse reflectance, between 4500 and 10000 cm-1. The original NIR spectra went through different transformations and mathematical pre treatments, such as the Kubelka-Munk transformation; multiplicative signal correction (MSC); spline smoothing and movable average, and the data were scaled by variance. The regression model permitted the determination of the moisture content of the raw coffee samples with a standard error of calibration (SEC) = 0.569 g.100 g -1; standard error of validation = 0.298 g.100 g -1; correlation coefficient (r) 0.712 and 0.818 for calibration and validation, respectively, and average relative error of 4.1% for validation samples.
A espectroscopia na região do infravermelho próximo (NIR) foi usada para determinar o teor de umidade em amostras de café cru. Foram construídos modelos de regressão usando o método dos mínimos quadrados parciais (PLS) com diferentes pré-tratamentos de dados e 157 espectros NIR coletados de amostras de café usando um acessório de reflectância difusa, na região entre 4500 e 10000 cm–1. Os espectros originais passaram por diferentes transformações e pré-tratamentos matemáticos, como a transformação Kubelka-Munk
a correção multiplicativa de sinal (MSC)
o alisamento com SPLINE e a média móvel, e os dados foram escalados pela variância. O modelo de regressão permitiu determinar o teor de umidade nas amostras de café cru com erro quadrático médio de calibração (SEC) de 0,569 g.100 g –1
erro quadrático médio de validação de 0,298 g.100 g –1
coeficiente de correlação (r) 0,712 e 0,818 para calibração e validação, respectivamente
e erro relativo médio de 4,1% para amostras de validação.
metadata.dc.description.abstractalternative: Near infra-red reflectance (NIR) spectroscopy was used to measure the moisture content in raw coffee. Different models using partial least squares (PLS) with data pre-processing were used. Regression models were built with 157 spectra of the samples of raw coffee collected using a near infrared spectrometer with an accessory of diffuse reflectance, between 4500 and 10000 cm–1. The original NIR spectra went through different transformations and mathematical pre treatments, such as the Kubelka-Munk transformation
multiplicative signal correction (MSC)
spline smoothing and movable average, and the data were scaled by variance. The regression model permitted the determination of the moisture content of the raw coffee samples with a standard error of calibration (SEC) = 0.569 g.100 g –1
standard error of validation = 0.298 g.100 g –1
correlation coefficient (r) 0.712 and 0.818 for calibration and validation, respectively, and average relative error of 4.1% for validation samples.
Subject: Umidade
Café
Country: Brasil
Editor: Sociedade Brasileira de Ciência e Tecnologia de Alimentos
Citation: Food Science and Technology (Campinas). Sociedade Brasileira de Ciência e Tecnologia de Alimentos, v. 28, n. 1, p. 12-17, 2008.
Rights: aberto
Identifier DOI: 10.1590/S0101-20612008000100003
Address: http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0101-20612008000100003&lng=en&nrm=iso&tlng=pt
Date Issue: 2008
Appears in Collections:FEA - Artigos e Outros Documentos

Files in This Item:
File Description SizeFormat 
S0101-20612008000100003.pdf477.96 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.