New Prototype of Hybrid 3D-Biometric Facial Recognition System

In the last decades, a lot of 3D face recognition techniques have been proposed. They can be divided into three parts, holistic matching techniques, feature-based techniques and hybrid techniques. In this paper, a hybrid technique is used, where, a prototype of a new hybrid face recognition technique depends on 3D face scan images are designed, simulated and implemented. Some geometric rules are used for analyzing and mapping the face. Image processing is used to get the twodimensional values of predetermined and specific facial points, software programming is used to perform a three-dimensional coordinates of the predetermined points and to calculate several geometric parameter ratios and relations. Neural network technique is used for processing the calculated geometric parameters and then performing facial recognition. The new design is not affected by variant pose, illumination and expression and has high accurate level compared with the 2D analysis. Moreover, the proposed algorithm is of higher performance than latest's published biometric recognition algorithms in terms of cost, confidentiality of results, and availability of design tools