Satellite images provide a wealth of visual data from which we can visualize in interesting ways. Land Lines is an experiment that lets you explore real satellite imagery through commonly used mobile gestures–drawing and dragging.
“Draw” finds satellite images that match each line you draw. “Drag” creates a line of connected rivers, coastlines, and roads that move in the direction of where you drag. The result is a unique and unexpected approach to exploring our planet. Land Lines was made by the media artist and developer Zach Lieberman and the Google Data Arts Team. Learn more about how the project was created in this technical case study or browse the open-source code on GitHub.
Line Detection helped to select the most interesting images
We used a combination of OpenCV Structured Forests and ImageJ’s Ridge Detection to analyze and identify dominant visual lines in the initial data set of 50,000+ images. This helped to reduce the original dataset to just a few thousand of the most interesting images.
UPDATE: Land Lines received the “FWA of the Day” on Dec 21, 2016