Articles
Traceability and fruit quality sensing on a platform harvester
Article number
1395_28
Pages
209 – 216
Language
English
Abstract
Temperate fruit tree industries in Australia are striving to improve production efficiency and packout yield to maximise profitability.
A critical component is to reduce the variability in fruit quality across an orchard block.
In addition, regulatory, biosecurity and consumer-driven product traceability are increasing the complexity of data requirements for fruit growers.
Advanced sensing systems and machine learning approaches offer non-destructive, real-time fruit quality appraisal.
A project to combine sensors, machine learning, orchard traceability and mechanisation into a system that captures fruit quality data when and where it is picked is underway at the Tatura SmartFarm.
The fruit quality assessment system comprises a fluorescence spectrometer, a reflectance spectrometer, an optical camera and a GPS system fitted in an enclosure attached to a conveyor arm on a platform harvester.
The system also features a WiFi modem enabling cloud connectivity.
Initially, optical imagery estimates of apple, pear, nectarine and plum fruit diameter, shape and skin colour were compared to hand-held calliper and colourimeter measures.
Calibration samples for size and colours were also employed and the results of the calibration are reported.
Field testing, including the validation of the fruit quality models to estimate starch index, ethylene emission, flesh firmness and soluble solids concentration is planned for season 2023-24. Non-destructive fruit quality sensing on the platform harvester, implications for traceability and improved precision orchard management are discussed.
A critical component is to reduce the variability in fruit quality across an orchard block.
In addition, regulatory, biosecurity and consumer-driven product traceability are increasing the complexity of data requirements for fruit growers.
Advanced sensing systems and machine learning approaches offer non-destructive, real-time fruit quality appraisal.
A project to combine sensors, machine learning, orchard traceability and mechanisation into a system that captures fruit quality data when and where it is picked is underway at the Tatura SmartFarm.
The fruit quality assessment system comprises a fluorescence spectrometer, a reflectance spectrometer, an optical camera and a GPS system fitted in an enclosure attached to a conveyor arm on a platform harvester.
The system also features a WiFi modem enabling cloud connectivity.
Initially, optical imagery estimates of apple, pear, nectarine and plum fruit diameter, shape and skin colour were compared to hand-held calliper and colourimeter measures.
Calibration samples for size and colours were also employed and the results of the calibration are reported.
Field testing, including the validation of the fruit quality models to estimate starch index, ethylene emission, flesh firmness and soluble solids concentration is planned for season 2023-24. Non-destructive fruit quality sensing on the platform harvester, implications for traceability and improved precision orchard management are discussed.
Authors
D. Pelliccia, M.G. OConnell, N. Valluri, A. Scalisi, I. Goodwin
Keywords
deep learning, fruit maturity, optical imagery, spectrometry
Groups involved
- Division Temperate Tree Nuts
- Division Vine and Berry Fruits
- Division Temperate Tree Fruits
- Division Precision Horticulture and Engineering
- Division Plant-Environment Interactions in Field Systems
- Division Tropical and Subtropical Fruit and Nuts
- Working Group Precision Management of Orchards and Vineyards
Online Articles (59)