Iris Identification enhancement based deep learning

Abstract

Iris  recognition   is one of the most  important  biometric solution, which is used for individuals identification because  of   its unique features;  so  always there is need to improve  and enhance recognition method used for that purpose,  the proposed   method  aim to  use  un enhanced iris detection methods to ensure iris detection accuracy with acceptable speed , for  large and different type of images from different source.  the data is process  using a set of  preprocessing steps  and  normalized ,  then using  deep learning technique which is  known as UNINET,  that work as feature extraction for the  iris part .The net also have  another  sub net which is responsible for detecting the non-iris  part that may  affected in recognition result, so iris and  non-iris part is depended during the matching stage. The proposed method gave promised result comparing with some of high benchmark methods, where the accuracy was 96.56, 99.26, and 99.8 for CASI, ITTD, and captured images correspondingly.