Researchers have improved the processing and compression of large hyperspectral imagery on cube satellites. This technology, using AI, allows earlier detection of bushfires from space, enabling quicker response times and potentially reducing loss of life and property.

A project funded by SmartSat CRC and led by the University of South Australia (UniSA) has developed an energy-efficient early fire smoke detection system for South Australia"s first cube satellite, Kanyini. The mission is a collaboration between the SA Government, SmartSat CRC, and industry partners to monitor bushfires and water quality using a 6U CubeSat satellite in low Earth orbit.

The satellite"s hyperspectral imager captures reflected light from Earth in various wavelengths to create detailed surface maps for applications like bushfire monitoring and land management.

Lead researcher Dr. Stefan Peters from UniSA noted that traditional Earth observation satellites lack the onboard processing to analyze complex images in real-time. His team, including scientists from UniSA, Swinburne University of Technology, and Geoscience Australia, developed a lightweight AI model to detect smoke within the cube satellite"s processing, power, and storage constraints.

The AI model reduced the data volume downlinked to 16% of its original size and used 69% less energy compared to ground-based processing. It detected fire smoke 500 times faster than traditional methods.

"Smoke is usually the first thing you can see from space before the fire gets hot and big enough for sensors to identify it, so early detection is crucial," Dr Peters says.

Using simulated imagery of recent Australian bushfires, the AI model was trained to detect smoke in an image. In a case study of a past fire event in the Coorong, the simulated Kanyini AI approach took less than 14 minutes to detect smoke and send data to the South Pole ground station.

"This research shows there are significant benefits of onboard AI compared to traditional on ground processing," Dr Peters says. "This will not only prove invaluable in the event of bushfires but also serve as an early warning system for other natural disasters."

The research team aims to demonstrate the AI fire detection system in orbit in 2025 with the Kanyini mission. They hope to commercialize the technology and use it in a CubeSat constellation to enhance early fire detection.

The experiment details are published in the latest issue of IEEE Journal of Selected Topics in Applied Earth and Remote Sensing.

Related Links
SmartSat CRC
Forest and Wild Fires - News, Science and Technology