Moisture content detection in organic substrates based on characteristic wavelength in near infrared spectroscopy
Key Laboratory of Modern Agricultural Equipment and Technology, Ministry of Education ＆ Jiangsu Province, Jiangsu University, Zhenjiang, Jiangsu 212013, China)
Abstract：For developing a method for rapid detection of moisture content in vinegar residue substrates, visible-near-infrared spectroscopy was used for collecting the spectral data of the representative samples. The moving average filter (MAF) and first derivative (FD) were selected as the spectra preprocessing method. In accordance with the change of determination coefficients, seven spectra sections were chosen for determining the characteristic wavelengths. Stepwise linear regression (SLR) method was applied to perform the calibration models for predicting the moisture content. The results indicated that the characteristic wavelength was at 1 844, 1 889, 1 544 and 1 735 nm, respectively. Based on these wavelengths, 5 linear regression models were performed by stepwise regression method. For comparing the predictive precision of these models, the correlationship between prediction moisture content and that obtained by ovendrying method was analyzed. It is confirmed that the multivariate equation with the characteristic wavelength 1 889, 1 544 and 1 735 nm had higher predictive precision with correlation coefficient 0.967 4 and root mean squared error of prediction (RMSEP) 0098 1. It is concluded that the visible-near-infrared spectroscopy with characteristic wavelength selecting is a useful method for rapid detection of organic substrate moisture content.
2011, Vol. 32 
