Automatic fruit sorting by non-destructive determination of quality parameters using visible-near infrared to improve wine quality: II. Regression analysis

In the last edition of NIR news1 we presented our research project which focuses on the determination of suitable parameters correlated with grape and wine composition which could be implemented as quality criteria for improving existing sorting methods for grapes. Whereas the last article focused on a description of the project, the sorting tests done during harvest, the classification results (botrytis, berry size, sugar concentration), some initial laboratory results for tannin concentration and the concentration of the total anthocyanin of Pinot Noir, we now focus on the regression analysis for determination of sugar concentration.