Lightning is one of the most mysterious and fascinating phenomena on the planet. Extremely powerful, but each hit only has enough energy on average to power an incandescent bulb for an hour. The exact mechanism that triggers a lightning strike is still not well understood. Yet it happens 45 times per second somewhere on the planet. While we won’t be able to gain a deeper scientific appreciation of lightning anytime soon, we can still capture it in various photographs through this project that takes advantage of computer vision.
machine learning to shoot the best lightning frames.
The creator of the project, [Liam], designed it as a tool for storm chasers and photographers so they could shoot long periods of time and not have to manually rewind their footage to extract frames with lightning bolts. The project borrows from a similar project, but this one adds Python 3 functionality and runs on a small netbook for easier field deployment. It uses OpenCV for object recognition, using video files as source data, and offers different modes to recognize different types of lightning.
The software is free and open source, and versions are supported for Windows and Linux. So far, [Liam] was able to capture all sorts of electrical atmospheric phenomena, including lightning, red sprites, and elves. We don’t see too many projects involving lightning here, in part because humans can only generate a fraction of the voltage potential needed for an average lightning strike.