ISHS Contact


ISHS Acta Horticulturae 1154: V International Symposium on Applications of Modelling as an Innovative Technology in the Horticultural Supply Chain - Model-IT 2015

Applying visible-near infrared (Vis-NIR) spectroscopy to classify 'Hayward' kiwifruit firmness after storage

Authors:   M. Li, R.R. Pullanagari, T. Pranamornkith, I.J. Yule, A.R. East
Keywords:   Actinidia, non-destructive, sorting, storability, multivariate data analysis
DOI:   10.17660/ActaHortic.2017.1154.1
A significant proportion of New Zealand's kiwifruit production is held as stock in local cool-stores for extended periods of time before being exported to global markets. The variability in fruit quality at harvest contributes to a wide range in fruit storage potential. The development of soft fruit (flesh firmness <9.81N) during storage and the difficulty in identifying and segregating them prior to distribution cause the industry financial loss. This study aimed to evaluate the feasibility of applying visible-near infrared (Vis-NIR) spectroscopy after cool storage, to segregate kiwifruit on their flesh firmness values so as to reduce the proportion of soft fruit for subsequent distribution in the supply chain. 'Hayward' kiwifruit (Actinidia deliciosa) from 51 growers were sourced from commercial orchards in New Zealand and stored in trays at 0C. At the end of storage (after 75, 100 or 125 days), fruit were scanned non-destructively using an NIR spectroradiometer in the reflectance mode before flesh firmness assessment. A global calibration model for all storage times was developed based on the spectral data using a support vector machine to categorise the fruit into two groups; 'Soft' (<9.81N) or 'Good' (≥9.81N). The validation model was able to accurately classify approximately 48% of soft fruit and 80% of good fruit irrespective of storage time. Applying the developed model for segregation before distribution would reduce the proportion of soft fruit from 25.4% in the original population to 17.9% in the remaining population after removal of 27% of the population (predicted as soft fruit). Sorting fruit by NIR after storage may enable reduction of the effort and cost to segregate poor fruit from good fruit throughout the supply chain.

Download Adobe Acrobat Reader (free software to read PDF files)

    1154     1154_2

URL      Hosted by KU Leuven      © ISHS