Enhanced Continuous Face Recognition Algorithm for Bandwidth Constrained Network in Real Time Application

Abstract

Current advancement in technologies enables improvements in terms of general welfare and security to be made. These days, biometric recognition is highly regarded as one of the safest and unique ways to verify, authenticate and provide access to the users. One known biometric recognition type is face. Face recognition (FR) is widely used for a number of reasons, ranging from verification purpose or to enabling access. While numerous studies on FR algorithm are carried out, there are still few researches that elaborate on the real-time application of the proposed algorithms. In this paper, the development of an FR algorithm meant for real-time application is described. The algorithm is developed based on the Discrete Krawtchouk Moment (DKM), which is known for its wide application in FR. Our work extends other work in this area by having an algorithm that is able to perform the recognition swiftly without consuming a lot of network bandwidth. Evaluation performed using two different types of camera of different resolutions confirms the ability of the proposed algorithm to fulfil its objectives.