Face recognition based on the phase spectrum of local normalized image

Research output: Contribution to conferencePaper

2 Scopus citations

Abstract

This paper proposes a robust faces recognition method based on the Phase Spectrum Features of the local normalized image. The Principal Components Analysis (PCA) and the Support Vector Machine (SVM) are used in the classification stage. We evaluate how the proposed method is robust to illumination, occlusion and expressions using "AR Face Database", which includes the face images of 109 subjects (60 males and 49 females) under illumination changes, expression changes and partial occlusion. The proposed method provides results with a correct recognition rate more than 95.5%. © 2008 IEEE.
Original languageAmerican English
Pages123-127
Number of pages5
DOIs
StatePublished - 29 Dec 2008
Event7th Mexican International Conference on Artificial Intelligence - Proceedings of the Special Session, MICAI 2008 -
Duration: 29 Dec 2008 → …

Conference

Conference7th Mexican International Conference on Artificial Intelligence - Proceedings of the Special Session, MICAI 2008
Period29/12/08 → …

    Fingerprint

Cite this

Olivares-Mercado, J., Hotta, K., Takahashi, H., Perez-Meana, H., & Sanchez-Perez, G. (2008). Face recognition based on the phase spectrum of local normalized image. 123-127. Paper presented at 7th Mexican International Conference on Artificial Intelligence - Proceedings of the Special Session, MICAI 2008, . https://doi.org/10.1109/MICAI.2008.46