Date of Award

Spring 2003

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Electrical & Computer Engineering

Program/Concentration

Electrical Engineering

Committee Director

Vijayan K. Asari

Committee Member

Ravindra P. Joshi

Committee Member

Frederic D. McKenzie

Call Number for Print

Special Collections LD4331.E55 G68 200

Abstract

This thesis describes research in automated methods for the recognition of human faces. The research is driven by the need to design a method, which would ensure high accuracy under the conditions of facial expression, illumination and pose variations. The resulting method is able to cope with uncontrolled nature of facial expression, illumination and head rotations. The main novelty of this work is the idea that some of the local facial features do not vary even when the facial expression, illumination and pose vary. This idea is applied to the existing principle component analysis lPCA) method to arrive at a new method called Modular PCA. The proposed face recognition algorithm is tested with three databases namely ODU, Yale and UMIST databases to evaluate its performance in the presence of variations in facial expressions, illumination and pose. The modular PCA technique is found to be more accurate than the conventional PCA method, with the highest improvement seen under the condition of varying illumination. A parallel VLSI architecture is also designed to realize the Modular PCA method in hardware. The architecture is simulated using Altera Quattus II and implemented on an APEX 20K series FPGA chip. The implementation is found to be ideal for real time applications.

Rights

In Copyright. URI: http://rightsstatements.org/vocab/InC/1.0/ This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).

DOI

10.25777/fyx4-3g36

Download

Share

COinS