Rotation Invariant Iris Recognition System for Person Identification

Author(s):

Sayali Lokhande , Sinhgad College of Engineering,Pune; S. R. Ganorkar, Sinhgad college of Engineering,Pune

Keywords:

Preprocessing Segmentation, Normalization, Random Transform

Abstract:

Iris based authentication system is pattern based system uses iris for person identification. Human Iris is a biometric that offers premium performance. Iris pattern is stable and most likely never changes throughout the person’s life. Main challenge in iris recognition is in case of images which are captured under non cooperative condition. Under rigid image capture conditions it is possible to obtain good quality images and achieve impressing accuracy, with very low error rates. This approach offers a fine analysis from global characteristics via the generated Radon profiles. The Radon transform has the ability to capture these orientations .A global structural similarity based on these directional features is then derived. In this proposed work our main goal is to develop the rotation-invariant good-quality set of features with compact representation without compromising on the accuracy. The novelty of proposed method is that it extracts the rotation invariant features using radon transform simulation studies is carried out to test the validity of the proposed algorithm. The results obtained are used to calculate CRR of the system.

Other Details:

Manuscript Id	:	IJSTEV5I1021
Published in	:	Volume : 5, Issue : 1
Publication Date	:	01/08/2018
Page(s)	:	72-75

Download Article