Throughout this spring semester, I worked in Dr. Spencer Lake’s Musculoskeletal Soft Tissue Laboratory to validate a Matlab-based script intended to determine three dimensional distances between points of interest using a stereovision camera written last semester. The importance of this 3D measurement system is to be able to quantify 3D strain in musculoskeletal soft tissues, like the ulnar collateral ligament of the elbow, which undergo multi-planar deformations in vivo. I worked closely with my mentor Leanne Iannucci to establish an efficient workflow in order to validate this Matlab script at different zoom settings on the stereovision camera. My objective was to create graphs correlating the location of the object being measured in the field of view with the error in measurement in order to see where the stereovision camera works and fails. In this workflow, the stereovision system was calibrated using the Matlab stereo camera calibrator application, stereo camera parameters were exported to the workspace, and the calibrated volume was visualized. A physical structure was then set up containing the calibrated volume as specified from the output of the Matlab script. With the calibration volume set up, several images of validation objects were taken both within and outside the boundaries of the box. Plots of distance from the center of the calibration volume versus error were created for each discrete camera zoom setting. At the end of this semester, graphs were established for the first zoom setting; due to COVID-19 data was not able to be gathered to generate any additional graphs. In the future, measurements will be taken on more zoom settings to gain an even clearer picture of the success of the stereovision camera.
Mechanical Engineering and Material Sciences Independent Study
Date of Submission
Shah, Vaishali; Iannucci, Leanne; and Lake, Spencer, "Validation of Stereovision Camera System for 3D Strain Tracking" (2020). Mechanical Engineering and Materials Science Independent Study. 128.