Multi triangle based automatic face recognition system by using 3D geometric face feature

摘要

Automatic human face recognition is required in numerous applications. In this paper, a new approach for automatic face recognition system is developed based on 3D geometric face features. The objectives of this system are generating the real time automatic personal identification system, improvement of fast and accurate face recognition. The system developed the algorithm for computing the accurate 3D measurement of face features. It is based on the perspective projection of a triangle constructed from three nodal points extracted from two eyes and lips corners. The set of non-linear equation is established the nodal points of a triangle built by any three points in a 2D scene and their corresponding projected image points. Some mathematical concepts are used to calculate the distances and angles between feature points. The proposed system is not only recognizing the person but also extracting the personal information of a person from database.