A Wearable Computer Vision System for Measuring Shoe Position and Orientation

This thesis presents, as a component of the BalanceAid project, the initial design and prototype of a self-contained wearable system capable of measuring the relative position and orientation of the wearer's shoes. This system is the first published work to use two shoe mounted cameras to measure the position and orientation of the shoes during the entire gait cycle in real time and using entirely on-board processing. The system consists of a camera, eight LED markers, and a single board computer mounted to each shoe. All of the data processing is performed on the shoe mounted computers providing a 6 DoF coordinate system transformation from one shoe to the other at a rate of 15 Hz. Using 6 healthy subjects, experiments were performed to characterize the performance of the system. The system's measurements were compared to those of an accurate commercially available system.