Sugar content prediction of Sunshine mango using NIR spectrometer
Abstract
This thesis explores a near-infrared (NIR) spectroscopy technique is to classify and predict the maturity of sunshine mango based on its total soluble solid (TSS). In this project, NIR spectrometer was used to obtain the reflectance wavelength of the sunshine
mango. The juice from the mango was obtained and measured with Brix refractometer to measure the actual TSS value. Then, the acquired NIR wavelength was analysed and correlate with the actual TSS value using multivariate analysis which is partial least squares (PLS) regression. The average multiple correlation regression (R²) and root mean square error for calibration and cross-validation in replication (0.983 and 0.111 respectively) and (0.385 and 0.633 respectively) were found which explained that there was a high correlation between the wavelength and TSS. The larger difference in R² for calibration and cross-validation might be caused by different maturity in each of replication or unstable of prediction. In pre-treatment data, smoothing was used to improve the result in replication.