By keeping image processing gated to a local network of cameras, the team from Griffith University have bypassed the traditional need to store sensitive data on a central system.
Their case study, published in Information, Technology & People this week, was undertaken at Gold Coast Airport, which prior to the COVID-19 pandemic was one of the busiest in the state of Queensland with about 17,000 passengers passing through its doors daily.
Professor Dian Tjondronegoro said data collection had long been a major concern with conventional technology because it had to constantly observe people's activities, which poses privacy risks.
In contrast, the AI system could generate heat maps instead of using individuals' images to show social distancing breaches which has become an essential requirement during the pandemic.
"Our goal was to create a system capable of real-time analysis with the ability to detect and automatically notify airport staff of social distancing breaches," Tjondronegoro said.
"With our system, the central machine only needs to periodically call on local nodes to send updates they've made to their decision-making models without needing to see the images they've captured."
The researchers also tested several cutting-edge algorithms, lightweight enough for local computation, to automatically monitor crowd counting in congested airport locations such as the food court and check-in and waiting areas.
Tjondronegoro is confident the AI system would be flexible enough to allow authorities to double check results, reducing data bias and improving transparency in how the system works.
"The system can scale up in the future by adding new cameras and be adjusted for other purposes," he said. "Our study shows responsible AI design can and should be useful for future developments of this application of technology."