Publications

Multispectral infrared face recognition: a comparative study

Moulay Akhloufi and Abdel Hakim Bendada

Abstract - Face recognition in the infrared spectrum has attracted a lot of interest in recent years. Many of the techniques used in infrared are based on their visible counterpart, especially linear dimensionality reduction techniques like PCA and LDA. In the infrared spectrum, variations can occur between face images of the same individual due to pose, metabolic, time changes, etc. In this work we introduce non linear dimensionality reduction techniques and a probabilistic Bayesian technique face recognition in the infrared spectrum. This techniques permit to reduce intrapersonal variation, thus making them very interesting for infrared face recognition. A comparative study is conducted in order to evaluate the performance of the proposed techniques for infrared face recognition. In this work, we introduce four non linear dimensionality reduction approaches for multispectral face recognition. For this purpose, the following techniques were developed: global non linear techniques (Kernel-PCA, Kernel-LDA) and local non linear techniques (Local Linear Embedding, Locality Preserving Projection). The performances of these techniques were compared to classical linear techniques for face recognition like PCA and LDA. Two multispectral face recognition databases were used in our experiments: Equinox Face Recognition Database and Laval University Database. Equinox database contains images in the Visible, Short, Mid and Long waves infrared spectrums. Laval database contains images in the Visible, Near, Mid and Longwaves infrared spectrums with variations in time and metabolic activity of the subjects. A new technique for infrared face extraction based on SVM learning and recognition of thermal/reflectance data is introduced. Also, a novel approach for infraredface image alignment is presented. This approach permits the extraction of eyes and mouth positions in order to align, normalize and extract the face images prior to learning and recognition. Experimental results show that the Bayesian technique is promising and lead to interesting results in the infrared spectrum when a sufficient number of face images is used in an intrapersonal learning process. Also, we see interesting recognition performances using local non linear dimensionality reduction techniques for infrared face recognition in the short wave infrared spectrum.

Bibtex:

@inproceedings{Akhloufi818,
author    = { Moulay Akhloufi and Abdel Hakim Bendada },
title     = { Multispectral infrared face recognition: a comparative study },
booktitle = { Proc. QIRT 10 – Quantitative Infrared Thermography },
pages     = { X-X },
address   = { Quebec, QC, Canada },
year      = { 2010 },
month     = { July },
language  = { English }
}

Last modification: 2010/03/14 by akhloufi

		Home \| About \| People \| Research \| Publications \| Events \| Profile
		©2002-. Computer Vision and Systems Laboratory. All rights reserved