Pests and diseases pose a significant production concern to most ginger growers and are a constant threat to yields. Fusarium, in particular, is a key threat to seed ginger stock, as it can be spread through soil from infected plant material. Identification and removal of diseased seed stock is currently performed manually and is a major production cost.
The preparation of disease-free seed ginger takes place over three months and is a labour-intensive process (e.g. production of 400t requires about 20 people full-time for 12 weeks per year). Additionally, the nature of the process requires operators to stand for long periods of time while visually inspecting and preparing the seed stock.
This project aims to develop and demonstrate an automated vision system that is capable of robustly identifying signs of Fusarium in seed ginger stock.