BloomFinder is developing new ways of tracking wildflower blooms using social media, computer vision, and remote sensing. We are teaching computers to recognize hundreds of common wildflowers in mountain ecosystems across dozens of sites in the Western United States. We aim to set these algorithms loose on publicly available Flickr and Instagram photos to automatically detect when and where blooms happen. This data will help us get a better handle on how plants are affected by extreme weather events, which are becoming more frequent under climate change.
BloomFinder is led by Ian Breckheimer and Charles Davis in the Department of Organismic and Evolutionary Biology at Harvard University. Support from the project comes from the National Science Foundation. Keep abreast of developments by following us on Twitter and Facebook.
Follow us on Twitter or Facebook for updates on how to contribute to BloomFinder.