Government trialling machine learning tech to detect pests at shipping ports

2Mar - by aiuniverse - 0 - In Machine Learning

Source – https://www.themandarin.com.au/

The federal government is working with a Canberra-based company to trial machine learning technology that aims to detect pests at Australian shipping ports.

The trial is being implemented in partnership with the Department of Industry, Science, Energy and Resources, the Department of Agriculture, Water and the Environment, Australian company Trellis Data, and global logistics company DP World, in Brisbane.

The technology will stop pests from breaching biosecurity processes at Australia’s container ports by detecting invasive species that are less than 10 millimetres in size, according to Trellis.

“The technology allows pests to be detected on the outside of every container that is transported from ship to shore in Australia. It also has the same detection capability for any internal container inspections required by biosecurity authorities,” the company told The Mandarin.

“The software works in all environments 24/7, identifying pests in real-time. Not only does it identify pests, it explains its reason for the identification and ensures all detections are correctly associated with the individual container ID.”

A recent CSIRO report found the amount of biosecurity risk materials intercepted in Australia increased by almost 50% in the five years to 2017. It warned that outbreaks across biosecurity sectors “are continuing to rise in volume and complexity”, and called for transformational changes to Australia’s biosecurity system.

The CSIRO noted that greater levels and speed of global trade would create new opportunities for pests and diseases to enter and spread across Australia.

Trellis said that initial progress has shown “very promising” results. The first official trial figures are expected to be released in May, with final evaluation to be completed by the end of the year.

Facebook Comments