A New Fast Facial Recognition Algorithm Applicable to Large Databases

Author

Abdelwahab, Moataz M. ; Mikhael, Wasfy B.

Author_Institution

Sch. of Electr. Eng. & Comput. Sci., Central Florida Univ., Orlando, FL

fYear

2006

fDate

38869

Firstpage

193

Lastpage

196

Abstract

In this contribution, a transform domain two-dimensional principal component analysis algorithm employing vector quantization (TD2DPCA/VQ) is presented for facial recognition, particularly for large databases. The algorithm has attractive properties with respect to storage requirements in the training mode and the computational complexity in the testing mode. The experimental results obtained by applying the new algorithm to the ORL database confirmed the significant reduction in the storage and computational requirements while improving the excellent recognition accuracy of the spatial 2DPCA method

Keywords

computational complexity; face recognition; principal component analysis; vector quantisation; visual databases; 2D principal component analysis; ORL database; computational complexity; facial recognition; transform domain; vector quantization; Computer science; Covariance matrix; Face recognition; Feature extraction; Image databases; Image recognition; Image storage; Principal component analysis; Testing; Vector quantization;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 2006 IEEE North-East Workshop on

Conference_Location

Gatineau, Que.

Print_ISBN

1-4244-0416-9

Electronic_ISBN

1-4244-0417-7

Type

conf

DOI

10.1109/NEWCAS.2006.250916

Filename

4016947