Prediction of Chufa MT-firmness Using FT-NIR Spectroscopy

Guang Ma,Yibin Ying,Xiaping Fu,Huishan Lu,Haiyan Yu,Yande Liu

DOI: https://doi.org/10.1117/12.630984

2005-01-01

Abstract:Chufa (Eleocharis tuberose Schult) is a special local product in south China. It is both vegetable and fruit. Near infrared spectroscopy was widely used for fruit and vegetable quality evaluation. The objective of this research was to study whether Chufa MT-firmness can be nondestructively measured by NIR technology and chemometrics methods. Two hundreds and thirty-nine samples were collected from two different cultivate regions and in each region three plots were chosen. NIR spectral data were acquired in the spectral region between 800 nm and 2500 nm using Nicolet FT-NIR spectrometer. Firmness was detected by a biomaterial universal testing machine. Chemometrics methods of PLS, PCR and SMLR were applied to establish statistical models for establishing the relationship between Chufa NIR spectra and MT-firmness in three different spectral regions of 800-2500 nm, 830-1250 nm and 860-1090 nm. The PLS model educed better results than PCR and SMLR models. And for the three spectral regions, the full spectral region of 800-2500 nm was better than other two. The correlation coefficient (r), root mean square error of calibration (RMSEC), root mean square error of prediction (RMSEP) and root mean square error of cross validation (RMSECV) of the PLS model in the range of 800-2500 nm were 0.74, 4.96 N, 5.63 N and 5.38 N respectively.