Real-Time Periocular Based Face Recognition System

This paper presents a stable technique for extracting periocular characteristics from photographs, with an emphasis on critical regions such as eye outlines and their environs. Our technique, which employs edge detection, Gabor filters, and Local Binary Patterns (LBP), accurately captures differentiating periocular properties required for security systems. It properly deals with issues including illumination differences, occlusions, and accessories. In addition, the system has a face identification component optimized for congested situations, which can recognize numerous faces and provide a count of those spotted. Experimental evaluations validate the algorithm’s efficacy, highlighting its potential for biometric authentication and monitoring. Detecting and counting faces in crowded environments provides vital information about crowd control, safety, and operational efficiency in a variety of scenarios.