Trellis Data, a prominent AI solutions provider, and the Australian Department of Agriculture, Fisheries, and Forestry have successfully concluded a pivotal five-month pilot trial of their Biosecurity Automated Threat Detection System (BATDS) at the DP World facility located at the bustling port of Brisbane. This momentous achievement marks a significant stride in fortifying Australia's $90 billion agriculture sector against the perils posed by invasive pests introduced through international trade.
The relentless threat of pests infiltrating the Australian agriculture sector through its ports necessitated a pioneering solution. The department entrusted Trellis Data with the task of crafting an AI-powered surveillance system capable of meticulously scrutinizing all incoming shipping containers. Leveraging the cutting-edge Trellis Intelligence Platform, the team devised custom Object Detection Models, seamlessly intertwining them with state-of-the-art camera management technology, ultimately birthing the innovative Biosecurity Automated Threat Detection System (BATDS).
The AI model, initially honed using Trellis Data's synthetic generation technology, exhibited remarkable evolution throughout the trial. It triumphantly identified approximately 58% of container anomalies detected through manual inspections, with a standout achievement of a 63% success rate in soil detection during the trial's culmination.
Tim McLaren, Head of Communication at Trellis Data, expressed his optimism, stating, "We believe this marks the initial stride towards incorporating AI into the Government's arsenal of surveillance tools. Expansions of this technology carry far-reaching implications for the Australian economy, transcending biosecurity, and encompassing goods protection, identification of illicit items, and broader border security."
Over the course of the pilot initiative, BATDS diligently scanned over 48,000 containers for biosecurity risk materials employing 35 strategically positioned cameras atop five DP World cranes. The system seamlessly integrated with existing procedures, ensuring efficient container scanning without hampering the flow of commerce. The department manually scrutinized 1,300 high-risk containers during the pilot, cross-referencing the results with BATDS detections.
A continuous refinement process, in collaboration with department entomologists, further fine-tuned the model. This optimization ensured more precise detections while effectively filtering out innocuous objects such as rust, grease, and container damages, ultimately culminating in a highly proficient machine learning model for biosecurity screening.
To facilitate comprehensive tracking of all detections, a separate model was developed to read container IDs on challenging surfaces. Trellis Data's ingenious system design allowed for wireless image streaming from the cameras to servers, facilitating seamless crane movement within the port.
Tim McLaren jubilantly declared, "This pilot project is a resounding success, underscoring the immense potential of our AI-driven system to amplify biosecurity screening on a grand scale. We've gained invaluable insights and lessons from this trial, and now it's time to apply our newfound knowledge on a larger canvas."
With ongoing enhancements and the momentum garnered from this groundbreaking AI-driven system, Trellis Data envisions a future where biosecurity screening across all port facilities can be substantially fortified. This advancement holds the promise of instilling greater confidence in the biosecurity status of containers entering Australia, thereby fortifying the country's crucial agriculture sector against external threats. Looking ahead, potential collaborations with other countries, such as the United States, are being explored to further validate and expand this transformative system.
###