An OpenCV-based computer vision system inspired by ReacTIVision.
PringleCV was designed for students of UQ's CSSE3010 - Embedded Systems Design & Interfacing course.
It provides the capability to track and localise a painted, labelled can (mounted on a rover vehicle) in 3D space using a single, low-cost webcam. Telemetry is available to external agents via UDP broadcast, a TCP server and a publish/subscribe topic on a Mosquitto server (MQTT).
PringleCV consists of 3 applications:
-
webcam.py- tracking via computer vision -
telemetry.py- distributing telemetry data -
webserver.py- 3D visualisation of the position using WebGL