Development of a Face Recognition System and Its Intelligent Lighting Compensation Method for Dark-Field Application

A face recognition system which uses 3D lighting estimation and optimal compensation for dark-field application is proposed. To develop the proposed system, can realize people identification in a near scene environment, light-emitting diode (LED) overhead light, eight LED wall lights, visible light binocular camera, control circuit are used. First, 68 facial landmarks detected their coordinates both image as well camera coordinate systems computed. Second, morphable model (3DMM) developed after considering shadows, transformation matrix between 3DMM estimated. Third, to assess uniformity, 30 evaluation points selected from face. Sequencing computations of radiation intensity, ray reflection luminance, response, uniformity then carried out. Ray occlusion processed using simplified model. Fourth, an realized: used flood lighting, lights employed meticulous lighting. genetic algorithm identify lights. Finally, Eigenface method recognition. The results show that our improve accuracy by >10% compared traditional methods.