An Near-infrared Prediction Model for Quality Indexes of Potato Processing
马铃薯加工品质指标近红外预测模型研究

200 different kinds of potato are taken as samples. The samples are preprocessed with the vacuum microwave freeze-dry technique. A model for predicting four main processing quality indexes (water, reducing sugar, starch and protein) of potato is established by using the Fourier transform near-infrared spectroscopy. By taking the determination coefficient (R2), root mean square error of cross validation (RMSECV), root mean square error of prediction (RMSEP) and relative predictive determination (RPD) as the accuracy evaluation indexes of the model, the prediction results of the model are verified with 50 kind-unknown potato samples externally. The external verification result shows that it is feasible to predict the main processing quality indexes of potato. The prediction effectiveness is ideal, but the prediction accuracy should be further improved for protein and reducing sugar.