Integration of face recognition and sound localization for a smart door phone system

This paper proposes a smart system using both face recognition and sound localization techniques to identify the faces of visitors from a door phone in much efficient and accurate ways. This system is effectively used to recognize the faces when their locations are out of the boundaries of the camera scope of the door phone. The smart door phone system proposed in this paper uses a visitor's voice source to rotate the camera to his face and then to recognize the face accurately. The integrated system has been designed with one FPGA(Field Programmable Gate Array) chip and tested for actual use in door phone environments.