The development of a face recognition system based on deep learning

Biometry is a science that allows people to be distinguished from each other by examining their physical and behavioral characteristics. Biometric systems, on the other hand, are established by examining the biometric properties of human beings to determine their identity. Face recognition is one of the fields of biometric systems and uses details such as eyes, nose, mouth to identify people. Increased concerns about public security have raised the need for authentication for physical and logical access. In addition, intelligent devices management and facial analysis and classification areas increase the interest in automatic facial recognition. Many studies have been conducted in the literature on the development of facial recognition systems. In this thesis, deep learning methods which are used day by day are mentioned and it is aimed to develop facial recognition system with these methods. Different sample sets were taken from the data set used and hyper-parameters were tested with these samples. Gender discrimination and facial recognition were performed using hyper-parameters that increased performance. Success in gender discrimination was improved by 98.25% and in face recognition by 96.25%. Keywords: Biometric Systems, Deep Learning, Face Recognition Systems.