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.

No abstract

No abstract

No abstract

Although audio-visual human systems have several well-known limitations, artificial sensors can measure information beyond our limits. What would happen if we were able to overcome our limitations? Would we be able to obtain a better knowledge of our environment? Or the information beyond our limits is redundant? In this paper, we compare infrared, thermal and visible images from an information theory point of view. We have acquired a small database and compared several measurements over these images. While infrasounds and ultrasounds are not directly applicable, for instance, to speaker recognition due to the impossibility of human beings generating sounds in these frequencies, this is not the case with image signals beyond the visible spectrum for face recognition. We have observed that visible, near-infrared and thermal images contain a small amount of redundancy (less than 1,55 bits).