Defense Date

2005

Document Type

Thesis

Degree Name

Master of Science

Department

Computer Science

First Advisor

Dr. Lorraine M. Parker

Abstract

Spatial databases store geometric objects and capture spatial relationships that can be used to represent key features of the human face. One can search spatial databases for these objects, and seek the relationships between them, using fuzzy logic to provide a natural way to describe the human face for the purposes of facial characterization. This study focuses on community perception of short, average, or long nose length. Three algorithms were used to update community opinion of nose length. All three methods showed similar trends in nose length classification which could indicate that the effort to extract spatial data from images to classify nose length is not as crucial as previously thought since community consensus will ultimately give similar results. However, additional testing with larger groups is needed to further validate any conclusion that spatial data can be eliminated.

Rights

© The Author

Is Part Of

VCU University Archives

Is Part Of

VCU Theses and Dissertations

Date of Submission

June 2008

Share

COinS