{"id":47,"date":"2025-04-22T05:11:33","date_gmt":"2025-04-22T12:11:33","guid":{"rendered":"https:\/\/scienceblog.com\/sciencechina\/?p=47"},"modified":"2025-04-22T05:11:33","modified_gmt":"2025-04-22T12:11:33","slug":"new-ai-drones-spot-victims-through-forest-darkness","status":"publish","type":"post","link":"https:\/\/scienceblog.com\/sciencechina\/2025\/04\/22\/new-ai-drones-spot-victims-through-forest-darkness\/","title":{"rendered":"New AI Drones Spot Victims Through Forest, Darkness"},"content":{"rendered":"

Finding survivors in wilderness after disasters has long been a race against time, with rescue teams battling challenging terrain and weather conditions. Now, researchers have developed an artificial intelligence system that dramatically improves how drones detect people during search and rescue missions\u2014even in dense forests or darkness.<\/p>\n

A team from Northwestern Polytechnical University and Yan’an University in China has created a dual-vision approach that combines regular camera footage with infrared imaging, allowing drones to spot victims that would otherwise be nearly invisible from above.<\/p>\n

“Our research contributes to the development of more effective Aerial Person Detection for search and rescue missions,” said Dr. Xiangqing Zhang, the study’s lead researcher. “By integrating AI with multimodal data fusion, we have designed a system that improves detection capabilities in complex environments, making SaR operations more efficient and reliable.”<\/p>\n

The technology addresses several longstanding challenges that have hampered aerial detection efforts. Traditional drone cameras struggle with people partially hidden by trees, lying motionless, or blending into their surroundings. Small objects\u2014like a person viewed from high altitude\u2014are particularly difficult for AI systems to recognize reliably.<\/p>\n

To solve these problems, the researchers built a custom unmanned helicopter equipped with synchronized visible and infrared cameras. This system can detect the heat signature of a person who might be invisible to regular cameras due to camouflage, darkness, or partial covering.<\/p>\n

The team compiled an extensive dataset called VTSaR featuring thousands of images captured across diverse environments\u2014from urban areas to wilderness and maritime settings. This dataset served as the foundation for training their detection algorithms, which achieved an impressive 95% accuracy in testing.<\/p>\n

What sets this research apart is the fusion of multiple AI approaches. The system combines object-aware methods to handle scale variations with information-fusion techniques that work across different lighting conditions. For rescue operations in remote areas with limited computing resources, lightweight algorithms ensure the system can operate efficiently on the drones themselves.<\/p>\n

Beyond search and rescue, this technology could transform other applications including disaster response, security monitoring, and wildlife conservation, offering a powerful example of how AI can enhance critical human safety operations.<\/p>\n

Read the Study<\/strong><\/p>\n