Personal Iris Recognition from an Image at Long Distance using Back Propagation Neural Network

Abstract

Now days, most reliable system for person identification is Iris recognition technique. Most of other systems are also presents for person identification like as identification cards or tokens, secret codes, passwords, etc. But the problems of these types of systems are, the secret codes and passwords can be cracked, the identification cards can be damaged. Therefore the effective method for the person identification is necessary. The iris recognition is treated as the most accurate method for personal identification. The person’s biometric physical and behavioral features are considered for the identification. This is most efficient technique, because the iris characteristics of a person cannot be change due to the age and environment. Therefore automatic systems are based on the iris recognition. In this paper, an effective technique for iris recognition is present to identify the individual. For the implementation of this system, the iris images are taken from UBIRIS.v1 database. This iris image is then segmented, normalized and features are extracted by using Hough Transform, Daugman rubber sheet modal and median filter. The matching of an iris is done with the help of back propagation neural network model. Also, the Chronological Monarch Butterfly Optimization -based Deep Belief Network (Chronological MBO-based DBN) is proposed for iris recognition to get better accuracy. All these operations are done on the MatLab.