Hyperspectral ZT and food safety determination (phase 2)

The objectives of this project were to analyse the spectral characteristics of contaminated meats using a hyperspectral camera, to develop a classification algorithm to discriminate between clean and contaminated meat and finally to design, construct, and trial a contaminant detection system in a red meat processing plant’s harvest room. 

As part of the project, lamb and contaminant samples were collected from a variety of sources. Preliminary samples were cut to contain equal amounts of fat and meat, and contaminants were applied in incremental amounts between scans to collect a diverse set of data.
Controlled tests showed that contaminants can be detected with an accuracy of more than 90 per cent.