A new scheme of signature extraction for iris authentication
This is an incremental improvement for high-security biometric systems, combining established techniques without novel breakthroughs.
The paper tackles iris authentication by developing a multi-stage signature extraction scheme using pupil extraction via grey level histogram, iris/eyelid edge detection via HOUGH transformation, iris encoding with 2D Gabor Wavelets, and authentication via Hamming distance.
Iris recognition, a relatively new biometric technology, has great advantages, such as variability, stability and security, thus is the most promising for high security environment. Iris recognition is proposed in this report. We describe some methods, the first one is based on grey level histogram to extract the pupil, the second is based on elliptic and parabolic HOUGH transformation to determinate the edge of iris, upper and lower eyelids, the third we used 2D Gabor Wavelets to encode the iris and finally we used the Hamming distance for authentication.