Color Iris Recognition Based on Quaternion

Spring 2010

Student
Miriam Cha
Advisor
Vijayakumar Bhagavatula
Project description

Personal identification is a critical and ubiquitous task in our daily lives. Conventionally, we prove our identities and gain access by physically carrying keys, ID cards or by remembering passwords. Unfortunately, conventional personal identification techniques are vulnerable; Physical items can be lost, stolen, or duplicated and passwords can be forgotten or compromised. Such deficiencies can be solved with biometric-based personal identification which is a method for uniquely recognizing humans based on one or more intrinsic physical or behavioral traits, such as fingerprints, iris, and face. Biometrics is a promising technology for improving security but it still has many technical challenges to overcome. The most common problem faced is object distortion due to time, aging, light conditions, and illnesses. In this project, I aim to improve the conventional iris recognition algorithm by developing a color iris recognition based on quaternion. We believe that color iris recognition using quaternion can provide improved iris recognition under challenging conditions.

Return to project list