What can’t satellites do? They keep an eye on animals, track humanitarian crises, even help predict famines—and their above-ground perspective lets ground-bound observers find unexpected beauty in their surroundings. Now, writes Eillie Anzilotti for CityLab, a new project is helping people find lovely patterns and strange similarities in cities around the globe.
Terrapattern launched last month, Anzilotti reports, and its concept is deceptively simple: Use satellite images to track specific visual features around large geographical regions. The open-source project uses machine learning to help people find places that look the same.
“We are particularly keen to help people identify, characterize and track indicators which have not been detected or measured previously,” the founders write, “and which have sociological, humanitarian, scientific, or cultural significance.”
That’s cool in theory, but addictive in practice. Users can pick a visual feature from one of thousands of high-res satellite images from five metro areas around the globe. A neural network then scans other cities for the same kinds of images.
Like baseball diamonds or Christmas tree farms? You can use Terrapattern to find a dizzying number of similar examples. But you don’t just have to look at recognizable objects like train tracks or runways—you can simply focus in on an area that has your favorite color or an interesting design.
The site was created in part by Golan Levin, a Carnegie Mellon art professor who’s obsessed with how humans interact with technology. His art does everything from help fonts evolve to add fingers to hands using creepy interactive software. He tells Anzilotti that he hopes the technology could be used to quickly locate disparities or environmentally damaging activities.
But for people obsessed with symmetry, visual style and the oddly satisfying, the project serves up so many soothing similarities that its potential benefits to the world are just a bonus. Can’t get enough? Head to Terrapattern’s stunning Flickr page for even more examples of Earth’s most gratifying patterns.