High-definition realistic image of hundreds of thousands of hectares of devices with increased content, modulating as a bar graph. The scenery captures vast landscape and intricate details depicting the contrast between nature and high tech indicators. The source of inspiration comes from online photographic content.

Intelligent Fire Detection: How HQPlantations is Using AI to Prevent Forest Fires

Uncategorized

Α forest fire can devastate thousands of hectares of land, leading to massive losses in timber resources and posing a threat to wildlife and human safety. HQPlantations, a leading timber company in Britain, is taking a proactive approach to fire management by using cameras, satellite imagery, and artificial intelligence (AI) to detect and predict fires within its vast plantation area of 288,000 hectares.

After a year-long trial period, HQPlantations entered into a commercial agreement with exci, a Sunshine Coast-based company now known as Fireball.International, to implement their fire management tool powered by AI. “During the trial, which took place during one of the calmest fire seasons in the last 20 years,” said Andrew Dunn, Fire Manager at HQPlantations, “the AI software from exci was able to detect fires based on smoke alone, proving its immense potential in fire detection.”

In addition to identifying fire events such as smoke and heat, exci uses machine learning models to predict when and where fires are likely to occur. These models are trained using more than 2.5 million camera images from their partners in Australia, North America, and South America, as well as 30GB of daily satellite data.

Dunn emphasized that while AI detection plays a crucial role, it is not the sole solution. The company will continue to rely on trained personnel and fire crews during periods of high fire risk and poor visibility caused by smoke. However, the AI-powered system will be invaluable during off-hours, particularly in the evenings and mornings when many fires start, often from abandoned vehicles.

The fire detection images provided by exci also aid in fire research, offering valuable insights into ignition time, single or multiple sources, and fire behavior. This information contributes to better understanding and management of fires in the future.

HQPlantations currently has cameras covering approximately 90% of its plantations, which are scattered throughout Queensland, from Passchendaele near the New South Wales border to Kuranda north of Cairns. “We have recorded several instances where exci’s detection has been the primary source for fire detection,” said Dunn.

To further enhance their fire management capabilities, HQPlantations maintains a fleet of 80 4×4 firefighting vehicles, including 12 tankers, and has more than 250 trained firefighters among their employees and partners. In case of a fire detection, exci sends notifications to the first response teams within two minutes using the Amazon Simple Notification Service and Amazon Simple Email Service.

Regarding the expansion of fire detection coverage, Dunn stated that the remaining 10% of plantations, particularly in Kalpowar in central Queensland and the southeastern region, are under evaluation for additional camera installations. As the fire risk increases in these areas, the company is considering further opportunities to leverage this technology.

Frequently Asked Questions