OPTIC FIBER SENSING TECHNIQUE FOR EVALUATING SOLUBLE SOLIDS CONTENT IN NAVEL ORANGE FRUIT USING VIS/NIR REFLECTANCE SPECTROSCOPY

摘要

The feasibility of Vis/NIR reflectance spectroscopic technology for rapid evaluation of soluble solids content (SSC) in navel orange fruit was investigated in the wavelength range of 350-1800nm. Partial least squares regression (PLSR) and back-propagation neural network (BPNN) based on principal component analysis (PCA) were used to develop the calibration models, with respect to multiplicative scatter correction ( MSC ) pretreatments, for predicting the SSC of intact navel orange fruit. The best combination, based on the prediction results for 38 unknown samples, was PCA-BPNN method resulting in correlation coefficient, root mean square error of prediction (RMSEP), average difference between predicted and measured values ( Bias ) of 0. 9034,0. 68280 Brix and 0. 15950Brix, respectively.Comparing the prediction results with conventional method, it can be found that the values between SSC measurement and PCA-BPNN prediction with MSC spectral pretreatment are not significantly different at 95 % confidence interval. Experimental results indicate that PCA-BPNN is a suitable tool to model the non-linear complex system, with additional advantages over PLSR, and the Vis/NIR spectrometric technique can be used for measuring the SSC of intact navel orange fruit,nondestructively.