International Journal of Advanced Research in Electronics and Communication Engineering (IJARECE) Volume 4, Issue 4, April 2015 741 ISSN: 2278 – 909X All Rights Reserved © 2015 IJARECE A novel approach towards iris segmentation and recognition Kalpana Singh, Dr. Rajat Gupta, and Dr. Kuldeep Pahwa Department of Electronics and Communication Engineering, Maharishi Markandeshwar University, Mullana (Ambala), Haryana AbstractIris segmentation is an integral part of iris recognition. Earlier works in the iris recognition uses the conventional iris localization and binarization to identify the pupil and iris boundaries. In this paper a new/ novel approach for iris segmentation is proposed where the existing canny method and local maxima suppressed image is used for identifying the boundaries and centre of pupil and iris.The paper shows the different result so obtained for iris segmentation using MATLAB software .CASIA database has been used to obtain the standardized output. Keywords: matlab, iris, casia, canny, maxima suppression 1. Introduction With increase in emphasis on security nowadays, biometric technologies are becoming much more important than ever [1]. In particular, iris recognition in recent years receives growing interests. Iris pattern recognition is unique to each subject,remain stable throughout life and offers several distinct advantages [2; 3; 1]. Especially, it is protected by the body’s own mechanisms and impossible to be modified without risk. Thus, iris is reputed to be the most accurate and reliable for person’s identification [5] and has received extensive attentions over the last decades.The degree of freedom of iris textures is extremely high, the probability of finding two identical irises is close to zero therefore, iris recognition systems are very reliable and could be used in most secure places. Iris segmentation is to locate the valid part of the iris for iris biometrics [7], including finding the pupillary and limbic boundaries of the iris, localizing its upper and lower eyelids if they occlude and detecting and excluding any superimposed occlusions of eyelashes, shadows or reflections. The centrality of segmentation to effectiveness of any iris recognition system cannot be overemphasized[4]. It determines effectiveness of the system [8].Two well-known iris segmentation approaches are attributed to Daugman and Wildes. Daugman developed integro-differential operator to find circular pupil and limbus boundaries. It can be interpreted as a circular edge detector, which searches, in a smoothed image by Gaussian filter, the parameters of a circular boundary along which the integral derivative is maximal [2]. Wildes proposed a two-stage iris segmentation method: a gradient based intensity image, and next the inner and outer boundaries are detected using Hough transform [9]. It is reported that most failures to match in iris recognition system result from inaccurate segmentation [10]. The contents of this paper are thus arranged: section 2 elucidates on available iris segmentation techniques. Section 3 identifies some available public iris databases that can be used for iris system validation. Section 4 gives an exhaustive literature review of iris segmentation methods discussed in literatures/researches while 5 and 6 discuss some limitations and areas of future researches respectively.