Biometric Verification using Periocular Features based on Convolutional Neural Network
S Srivika, L Gayathri, B Nivetha, Sri Devi N, Rushmitha Varshini Yuvaraj Sujatha · 2023
Periocular recognition is a biometric identification technology that uses the unique patterns and features of the skin surrounding the eye, such as the eyelids, eyelashes, and eyebrows, to recognize individuals. It is considered a subset of facial recognition technology, as the periocular region is a critical component of the face and contains important distinguishing features. Periocular recognition is becoming increasingly popular due to its high accuracy and usability in a variety of applications, including security and law enforcement, healthcare, and financial services. The process of periocular recognition involves capturing a digital image or video of the periocular region, extracting unique features using computer algorithms, and comparing these features to a pre-existing database of known individuals to identify a match. The network is trained on a collection of periocular pictures and their related IDs in order to employ Xception CNN for periocular identification. The network gains the ability to extract identifying characteristics from these photos that are discriminative. Experimental shows that the Xception CNN provide promising results in the recognition of periocular images.