Summary

The Australian ginger industry is a significant agricultural industry in Queensland generating $32 million gross volume for fresh ginger sales and $50 million semi-processed ginger value. The AgriFutures Ginger Program is looking to increase ginger production by 50% and the key to achieving this is technology to reduce disease impacts on yield and reduce labour inputs. The most common disease threat is Fusarium yellow rot. Growers consider a 10% loss in a patch due to disease acceptable, but at times some patches experience over 30-80% losses. Decreased yield has a financial impact on growers along with high labour requirements to detect disease in seed, with estimates of 400 tonnes requiring 20 people full time 12 weeks per year to process.
To address both disease detection and labour input issues, this project will use multispectral technology with a database model to enable instant detection of Fusarium in ginger rhizomes prior to planting. The advantage of multispectral imaging technology is the ability to detect what cannot be seen with the human eye combined with AI and machine learning to instantly detect Fusarium in ginger rhizomes.
Upon commercialisation via the patented technology & data can be accessed via a smart phone, handheld scanner, fixed scanner or conveyor belt attachment for flexible implementation by growers.
The team will deliver the project outcomes by working with ginger growers in Queensland, on farm and in laboratory setting, to ensure outcomes are applicable to real-world setting.

Program

Ginger

Research Organisation

Zondii

Objective Summary

This project contributes towards Priority three of the AgirFutures Ginger Program R&D Plan by exploring technologies new to the ginger industry, implementing a big data solution as well as delivering new and improved tools. The project will achieve this by introducing Multispectral imaging technology to the ginger industry coupled with a database solution and an easy to use tool for growers to use when sorting seed for planting. The ultimate goal of the research is to assist with automation and reduce labour costs.
The overall goal of this project is to apply the patented Zondii multispectral technology (www.zondii.com) to disease detection in the rhizomes of ginger. This will lead to improved output and quality in the ginger production industry and decreased labour hire costs.
This goal will be achieved by delivering the following objectives within 18 months:

Detect disease at a rate more accurate than the human eye due to additional information visible with multispectral imaging
Provide database with machine learning that detects ginger Fusarium and rhizome disease
Provide a smartphone prototype that detects diseased ginger seed.

We will achieve our objective through use of patented realtime scanning solution in a laboratory and real world setting to create a database of diseased and disease free rhizomes. AI and machine learning will be applied to the database to create a high precision testing tool with accurate and instant results.