Our new heat recovery tool uses AI-enhanced satellite and aerial imagery to help cities cope with dangerously high temperatures (and reduce associated health impacts). expected). The tool combines AI object detection with other models to analyze factors such as how much green space a city has and how well building roofs reflect sunlight. This will help urban planners and local governments understand which cooling interventions, such as tree planting, are likely to have the most impact on temperatures down to the neighborhood level. The tool is initially being piloted in 14 U.S. cities, where officials are using it to determine which areas are most at risk from heatwaves and develop support plans. As temperatures rise globally, AI could be part of the solution, especially for people living in urban heat islands.
Google Research partnered with leaders in the fire community to build FireSat, a constellation of satellites specifically designed to detect and track wildfires as small as a classroom (approximately 5 x 5 meters). FireSat gives authorities access to high-resolution imagery that is updated around the world every 20 minutes, allowing them to respond to fires as quickly as possible. This solves a critical problem in firefighting operations. Until now, firefighters had to use low-resolution satellite imagery, or satellite imagery that updated only a few times a day, making it difficult to spot fires until they were larger than a football field. Once a fire breaks out, emergency responders will receive this critical information at a time when it could be easier to suppress the fire.
We have made the bioacoustic-based model Health Acoustic Representations (HeAR) available to researchers. HeAR is designed to allow researchers to build models that can listen to human sounds, such as coughing, and signal early signs of illness. This technology could be useful in screening for diseases such as tuberculosis. Although tuberculosis is treatable, millions of cases go undiagnosed each year. Much of this is due to people not having easy access to health services, especially in countries without sufficient health resources. This kind of AI might help. For example, an India-based respiratory healthcare company is exploring how HeAR can improve early detection of tuberculosis based on cough sounds.
