M.S. Thesis Presentation by Meghan Shilling
Tuesday, December 17, 2002

(Dr. Thomas Kurfess, advisor)

"Two Dimensional Analysis of Mesoscale Parts Using Image Processing Techniques"

Abstract

With the increased precision of machines and processes, many devices are becoming smaller. As manufacturing processes output smaller and smaller parts, it becomes more critical that parts are measured on a smaller scale. This thesis focuses on analysis of mesoscale parts using commonly used measurement techniques, although many of the methods and results discussed can be applied to parts of all scales.

The main goal of this thesis is to develop a method to locate edge points from an initial three dimensional point cloud from a single surface. In order to achieve this goal, an image is created from the point cloud and image processing techniques are used to extract the edge points. Once the edges points have been located, analysis can be performed by fitting circles, lines, parallel lines, and ellipses to the edge points. A set of equations that can be used to fit these two dimensional curves with a minimizer is outlined.

Results from tests of these methods using both generated data and data from real parts are presented.