Segregation of ‘Hayward’ kiwifruit for storage potential using Vis-NIR spectroscopy
Loading...
Date
2022-07
Open Access Location
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Elsevier BV
Rights
(c) 2022 The Author/s
CC BY-NC-ND 4.0
CC BY-NC-ND 4.0
Abstract
Kiwifruit are often harvested unripe and kept in local coolstores for extended periods of time before being marketed. Many pre-harvest factors contribute to variation in fruit quality at harvest and during coolstorage, resulting in the difficulty in segregating fruit for their storage potential. The ability to forecast storage potential, both within and between populations of fruit, could enable segregation systems to be implemented at harvest to assist with inventory decision making and improve profitability. Visible-near infrared (Vis-NIR) spectroscopy is one of the most commonly used non-destructive techniques for estimation of internal quality of kiwifruit. Whilst many previous attempts focused on instantaneous quantification of quality attributes, the objective of this work was to investigate the use of Vis-NIR spectroscopy utilised at harvest to qualitatively forecast storage potential of individual or batches of kiwifruit. Commercially sourced ‘Hayward’ kiwifruit capturing large variability of storability were measured non-destructively at harvest using Vis-NIR spectrometer, and then assessed at 75, 100, 125 and 150 days after coolstorage at 0 °C. Machine learning classification models were developed using at-harvest Vis-NIR spectral data, to segregate storability of kiwifruit into two groups based on the export FF criterion of 9.8 N. The best prediction was obtained for fruit stored at 0 °C for 125 days: approximately 54% of the soft fruit (short storability) and 79% of the good fruit (long storability) could be predicted. Further novelty of this work lies within an independent external validation using data collected from a new season. Kiwifruit were repacked at harvest based on their potential storability predicted by the developed model, with the actual post-storage performance of the same fruit assessed to evaluate model robustness. Segregation between grower lines at harvest achieved 30% reduction in soft fruit after storage. Should the model be applied in the industry to enable sequential marketing, significant costs could be saved because of reduced fruit loss, repacking and condition checking costs.
Description
Keywords
Actinidia deliciosa, Storability, Near infrared, Prediction, Machine learning
Citation
Li M, Pullanagari R, Yule I, East A. (2022). Segregation of ‘Hayward’ kiwifruit for storage potential using Vis-NIR spectroscopy. Postharvest Biology and Technology. 189.