Engineer Frank Zhao has released a tool that allows OpenMV owners to put their cameras to work in the field of astrophotography, using computer vision to take some of the pain out of using polar scopes.
"Most of these [star] trackers or [astrophotography camera] mounts come with something called a polar-scope," Zhao explains. "A mini-telescope with a crosshair that you point at Polaris. The polar-scope doesn't just have a crosshair, it has a special circle reticle that requires you to take into account the current date, time, and location."
"You have to calculate the current sidereal time and adjust for your location's longitude to figure out where Polaris should be on the reticle circle. If I really wanted high precision, there are digital cameras that can help me polar-align. You attach the camera to your mount/tracker, calibrate it, point it roughly at Polaris, and they'll tell you how to adjust your mount to achieve polar-alignment."
The only problem: Said cameras are expensive and require the user to lug a laptop into the field. "At this point," Zhao explains, "I decided that they are over-priced for a glorified USB webcam with a CCTV lens, and I can make a better device myself."
That "better device" ended up being an OpenMV board, an all-in-one device designed for computer vision projects. Zhao's approach uses a plate-solving algorithm tailored to the OpenMV's built-in processor, meaning it doesn't require a laptop to work, while a Wi-Fi connection provides a user interface compatible with any browser — meaning the system can be used from a smartphone.
Zhao's creation has been released on GitHub under the permissive MIT License, with full instructions — and a detailed write-up on its creation.