A new face database simultaneously acquired in visible, near infrared and thermal spectrum

TL;DR

This paper introduces a comprehensive face database captured simultaneously in visible, near infrared, and thermal spectra, and demonstrates that combining these spectral data significantly improves face recognition accuracy across various illumination conditions.

Contribution

The paper presents a new multi-spectral face database and shows that combining visible, near infrared, and thermal images enhances recognition performance compared to single-spectrum systems.

Findings

Combining three spectral bands yields nearly equal contribution to recognition accuracy.

Significant improvement in identification rates when using combined spectra.

Achieved over 98% accuracy in six out of nine scenarios with a trained combination rule.

Abstract

In this paper we present a new database acquired with three different sensors (visible, near infrared and thermal) under different illumination conditions. This database consists of 41 people acquired in four different acquisition sessions, five images per session and three different illumination conditions. The total amount of pictures is 7.380 pictures. Experimental results are obtained through single sensor experiments as well as the combination of two and three sensors under different illumination conditions (natural, infrared and artificial illumination). We have found that the three spectral bands studied contribute in a nearly equal proportion to a combined system. Experimental results show a significant improvement combining the three spectrums, even when using a simple classifier and feature extractor. In six of the nine scenarios studied we obtained identification rates higher or equal to 98%, when using a trained combination rule, and two cases of nine when using a fixed rule.