Nondestructive measurement and fingerprint analysis of soluble solid content of tea soft drink based on Vis/NIR spectroscopy

A rapid, non-contact and nondestructive way to measure soluble solids content (SSC) of tea soft drink was put forward based on Vis/NIR spectroscopy, and the relationship between nondestructive Vis/NIRS measurement and SSC was developed. A field spectroradiometer was used with the wavelength from 325 to 1075 nm. The Vis/NIR spectra were acquired from 200 samples of tea soft drink. The multivariable analysis concluding partial least squares (PLS) and multiple linear regression (MLR) were applied to build the regression models and seek the fingerprint spectra. The excellent model with high prediction correlation (0.981) was obtained by PLS model based on entire wavelengths. The loading weights of latent variables from PLS model were used to look for the sensitive wavelengths. To evaluate the virtue of these wavelengths, the MLR models were built with these wavelengths. The significance of these wavelengths was analyzed by ANOVA, and five wavelengths (490, 498, 554, 929 and 970 nm) were selected with statistical significance at 99% confidence level. The acceptable result was obtained from MLR model with high prediction correlation (0.975). The potential of Vis/NIRS to predict SSC was ascertained, and the fingerprint spectra, which were strongly related with SSC, were obtained through multivariable analysis. These individual fingerprint wavelengths could be used to exploit a simple, low-cost and efficacious instrument.