BACKGROUND Visible-near infrared spectrometry is a technique suitable to assess chemical and physiological properties of fruit. Some models of calibration/prediction have been tested in order to assess the feasibility of a visible-near infrared sensor in order to monitor persimmon fruits colour, firmness, soluble solids, titratable acidity and soluble tannins. RESULTS Five regression models were investigated: principal component, partial least-squares, stepwise, support vector machines and ensembles of trees. These models were assessed by a 10-fold cross validation with a new strategy for both outliers removal and wavelengths reduction; furthermore, their statistical significance was evaluated by 100 Monte Carlo simulation runs. Principal component regression allowed to build excellent and/or very good fit/prediction models. The results (in terms of RPD as standard deviation to performance standard error ratio) are: 9.23 (+/− 0.26) for colour index, 10.18 (+/− 0.37) for firmness, 7.15 (+/− 0.28) for soluble solids content, 7.87 (+/− 0.31) for titratable acidity and 8.91 (+/− 0.33) for soluble tannins content. CONCLUSIONS The proposed strategy, for outliers removal and wavelengths reduction, allowed the achievement of useful results. Principal component regression fit/prediction capability resulted excellent. Conversely partial least-squares regression showed fair/poor results and the remaining tested models performed bad on real data.

Models to improve the non-destructive analysis of persimmon fruit properties by VIS/NIR spectrometry

ALTIERI, Giuseppe;GENOVESE, FRANCESCO;TAURIELLO, ANTONELLA;DI RENZO, Giovanni Carlo
2017

Abstract

BACKGROUND Visible-near infrared spectrometry is a technique suitable to assess chemical and physiological properties of fruit. Some models of calibration/prediction have been tested in order to assess the feasibility of a visible-near infrared sensor in order to monitor persimmon fruits colour, firmness, soluble solids, titratable acidity and soluble tannins. RESULTS Five regression models were investigated: principal component, partial least-squares, stepwise, support vector machines and ensembles of trees. These models were assessed by a 10-fold cross validation with a new strategy for both outliers removal and wavelengths reduction; furthermore, their statistical significance was evaluated by 100 Monte Carlo simulation runs. Principal component regression allowed to build excellent and/or very good fit/prediction models. The results (in terms of RPD as standard deviation to performance standard error ratio) are: 9.23 (+/− 0.26) for colour index, 10.18 (+/− 0.37) for firmness, 7.15 (+/− 0.28) for soluble solids content, 7.87 (+/− 0.31) for titratable acidity and 8.91 (+/− 0.33) for soluble tannins content. CONCLUSIONS The proposed strategy, for outliers removal and wavelengths reduction, allowed the achievement of useful results. Principal component regression fit/prediction capability resulted excellent. Conversely partial least-squares regression showed fair/poor results and the remaining tested models performed bad on real data.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11563/126052
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 18
  • ???jsp.display-item.citation.isi??? 14
social impact