The impact of collarette region-based convolutional neural network for iris recognition

Iris recognition is a biometric technique that reliably and quickly recognizes person by their iris based on unique biological characteristics. has an exceptional structure it provides very rich feature spaces as freckles, stripes, coronas, zigzag collarette area, etc. It many features where its growing interest in lies. This paper proposes improved method for identification Convolutional Neural Networks (CNN) with rate contribution area - the surrounding pupil recognition. Our work field of biometrics especially recognition; using full circle was compared detection lower semicircle collarette. The classification Alex-Net model to learn this feature, use couple (collarette/CNN) allows noiseless more targeted characterization also automatic extraction region, finally, SVM training used grayscale eye image data taken from (CASIA-iris-V4) database. experimental results show our proves be best accurate, because CNN can effectively extract higher accuracy new method, which uses achieved highest old methods region.