A NOVEL FACE RECOGNITION TECHNIQUE BASED ON COLOR AND EDGE FEATURES

Abstract

In current age of the internet, numerous services and apps have gone online, and each service is provided based on their identity management methods. This technique requires a user id and a strong password that cannot be broken by a standard brute force assault. However, maintaining a large number of passwords secret is a difficult effort, hence bio-matric authentication systems are used in a variety of critical environments to prohibit suspected or malevolent individuals. As a result, the various bio-matric authentication methods are evaluated first, and it is chosen to work with a face-based authentication system. In such systems, accuracy is the key criterion for system development, while resource preservation is the secondary concern of the face recognition system. Therefore, the goal of the proposed work to identify a face picture identification method that was both accurate and efficient. Accuracy, error rate, time complexity, and space complexity are utilised as performance metrics. Because of its high degree of accuracy in identifying user faces, the suggested model can be adopted. Additionally, the performance yields an image categorisation accuracy of 90–100%. As a result, the suggested face recognition method is effective and suitable for both object and face identification applications.