Designing a low-cost measurement system for volume estimation

Abstract

In this thesis, I designed and implemented large parts of a low-cost 3D reconstruction system targeted at volumetric reconstruction for lymphedema diagnosis purposes. The goal is to construct a system capable of determining changes in volume of an arm. This thesis involves construction of multi-camera rigs, but focuses on the development of data processing and surface reconstruction pipelines necessary to convert camera images into 3D surface models. We show that the system is capable of measuring volumes that, though error prone, are strongly correlated with true volume, and hence can detect volume changes on the order of 20 - 30%.