Algorithm for person identification by facial images
Shavkat Kh. Fazilov, Nomaz M. Mirzaev, Sobirjon Radjabov · ICEIC : International Conference on Electronics, Informations and Communications · 2008
a new algorithm for person identification by facial images is proposed in this paper. This algorithm is based on the method of face clustering. Algorithm has five stages: face localization, extracting of central part of a face, clustering, extraction of characteristic features and identification. Algorithm was tested with different test data and results of experiments show that proposed algorithm has advantages on other well-known methods in speed and accuracy.