Comparative analysis for a real time face recognition system using raspberry Pi

Abstract

Security is a major threat to institutions that is why there is a need of several specially trained personnel to attain the desired security to overcome the declining security conditions in the country. These personnel, as human beings, make mistakes that might affect the level of security. The need for facial recognition system that is fast and accurate is continuously increasing which can detect intruders and restricts them from restricted or high-security areas in real time and help in minimizing human error. Face recognition is one of the most important biometrics pattern recognition technique which is used in a broad spectrum of applications. The time and accuracy factor is considered as a major problem that specifies the performance of automatic face recognition system in real time environments. Various solutions have been proposed using multicore systems. However, harnessing current advancements is not without difficulties. Motivated by such challenge, this paper provides the architectural design, detailed design and proposes a comparative analysis for a Real Time Face Recognition System with three variant implementations of Real Time Face Recognition algorithms including Local Binary Patterns Histograms (LBP), PCA (Principal Component Analysis) and Fisher face. Finally, this paper concludes the speed obtained for the advanced implementations achieved by integrating embedded system models against the convention implementation.