Using chemometrics to characterise and unravel the near infra-red spectral changes induced in aubergine fruit by chilling injury as influenced by storage time and temperature
| Autor: | Farahmand Babellahi, Giancarlo Colelli, Leonarda Mastrandrea, Maria Lucia Valeria de Chiara, Federico Marini, Muhammad Mudassir Arif Chaudhry, Maria Luisa Amodio |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Materials science
rDCV Soil Science Near infra red Eggplant PLS-DA 01 natural sciences Chemometrics Component analysis Chilling injury ASCA NIR spectroscopy Spectroscopy Electrolyte leakage 010401 analytical chemistry 04 agricultural and veterinary sciences 0104 chemical sciences Horticulture Control and Systems Engineering 040103 agronomy & agriculture 0401 agriculture forestry and fisheries Classification methods Agronomy and Crop Science Food Science |
| Zdroj: | Biosystems Engineering. 198:137-146 |
| ISSN: | 1537-5110 |
| Popis: | The early non-destructive detection of chilling injury (CI) in aubergine fruit was investigated using spectroscopy. CI is a physiological disorder that occurs when the fruit is subjected to temperatures lower than 12 °C. Reference measurements of CI were acquired by visual appearance analysis, measuring electrolyte leakage (EL), mass loss and firmness evaluations which demonstrated that even before three days of storage at 2 °C, the CI process was initiated. An ANOVA-simultaneous component analysis (ASCA) was used to investigate the effect of temperature and storage time on the Fourier transform – near infra-red (FT-NIR) spectral fingerprints. The ASCA model demonstrated that temperature, duration of storage, and their interaction had a significant effect on the spectra. In addition, it was possible to highlight the main variations in the experimental results with reference to the effects of the main factors, and with respect to storage time, to discover any major monotonic trends with time. Partial least squares-discriminant analysis (PLS-DA) was used as a supervised classification method to discriminate between fruit based on chilling and safe temperatures. In this case, only significant spectral wavebands which were significantly influenced by the effect of temperature based on ASCA were utilised. PLS-DA prediction accuracy was 87.4 ± 2.7% as estimated by a repeated double-cross-validation procedure (50 runs) and the significance of the observed discrimination was verified by means of permutation tests. The outcomes of this study indicate a promising potential for near infra-red spectroscopy (NIRS) to provide non-invasive, rapid and reliable detection of CI in aubergine fruit. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect