Semester

Fall

Date of Graduation

2005

Document Type

Thesis

Degree Type

MS

College

Statler College of Engineering and Mineral Resources

Department

Lane Department of Computer Science and Electrical Engineering

Committee Chair

Natalia A. Schmid.

Abstract

Of the many biometrics that exist, iris recognition is finding more attention than any other due to its potential for improved accuracy, permanence, and acceptance. Current iris recognition systems operate on frontal view images of good quality. Due to the small area of the iris, user co-operation is required. In this work, a new system capable of processing iris images which are not necessarily in frontal view is described. This overcomes one of the major hurdles with current iris recognition systems and enhances user convenience and accuracy. The proposed system is designed to operate in two steps: (i) preprocessing and estimation of the gaze direction and (ii) processing and encoding of the rotated iris image. Two objective functions are used to estimate the gaze direction. Later, the off-angle iris image undergoes geometric transformations involving the estimated angle and is further processed as if it were a frontal view image. Two methods: (i) PCA and (ii) ICA are used for encoding. Three different datasets are used to quantify performance of the proposed non-ideal recognition system.

Share

COinS