Comparison of Masked and Unmasked Face Recognition Performance with Hand-Crafted Methods

Face recognition, a biometric technology, identifies people by their distinctive facial attributes. An image of a person's face must be compared to a library of other people's images to authenticate their identification. Face recognition systems may be used for identification, surveillance, and safety. Deep learning algorithms provide precise face recognition even under challenging conditions. Due to COVID-19 and masks, facial identification from unconstrained images is almost impossible. To avoid COVID-19, most people use masks outside. In many cases, typical facial recognition technology is useless. The majority of contemporary advanced face recognition methods are based on deep learning, which primarily relies on a huge number of training examples. Considering simply the eye and forehead regions of the face, however, masked face recognition may be investigated using hand-crafted approaches at a lower computing cost than using deep learning systems. In this thesis, handcrafted techniques are used to extract eye region and forehead characteristics from masked faces. We intend to construct a low-cost system for recognizing masked faces and compare its performance to that of face recognition systems that do not use masks. This study compares the performance of masked and unmasked face recognition systems. Experiments are undertaken on two publicly accessible datasets for masked face recognition: Masked Labeled Faces in the Wild (MLFW) and Cross-Age Labeled Faces in the Wild (CALFW). A comparison of the performance of the systems are provided in the thesis.