Achieving Information Security by multi-Modal Iris-Retina Biometric Approach Using Improved Mask R-CNN

The need for reliable user recognition (identification/authentication) techniques has grown in response to heightened security concerns and accelerated advances networking, communication, mobility. Biometrics, defined as the science of recognizing an individual based on his or her physical behavioral characteristics, is gaining a method determining individual's identity. Various commercial, civilian, forensic applications now use biometric systems establish purpose this paper design efficient multimodal system iris retinal features assure accurate human improve accuracy using deep learning techniques. Deep models were tested retinographies images acquired from MESSIDOR CASIA-IrisV1 databases same person. Iris region was segmented image custom Mask R-CNN method, unique blood vessels person principal curvature. Then, order aid precise recognition, they optimally extract significant information retina. suggested model attained 98% accuracy, 98.1% recall, precision. It been discovered that approach Iris-Retina improves efficiency recognition.