Inspiration Watching birds make aerial maneuvers at unknown speed and distance inspired us to devise a simple method for tracking motion. We were guided by a knowledge of ball-tracking systems in sports, but hoped to develop a simple, inexpensive, and portable version.
What it does The tracker takes video feed from two cameras and uses some light computer vision and math to triangulate a position of an identified object. The software then returns data on the motion of the object, in the form of trajectory plots and velocity/acceleration data, either at the end of a recording period or in real time.
How I built it We used openCV for object identification for each camera, wrote some numerical calculations to use the two sets of position data (one from each camera) to find position over time, and calibrated the device for proper alignment and accuracy. The calibration involved both physical fine tuning of alignment and numerical fitting of test data to find unknown camera parameters, such as FOV.
Challenges I ran into Calibration is an arduous process.
Accomplishments that I'm proud of It works! We functioned as a team very well, with every member playing a key role in the final result.
What I learned We developed collaborative skills and learned about the strengths and pitfalls of color-mask tracking.
What's next for SPACE Cam Take it out for applications! Tracking sports plays, wildlife, experiments in applied math...and who knows what else.