Multispectral images are images with more than one channel acquired in different bands or spectral ranges of the electromagnetic spectrum. Each one has specific details that can be exploited in facial recognition applications. In particular, to detect facial expression variations, pose variations and presentation attacks, a facial analysis system can benefit not only of images from the visible spectral band but also of infrared images. In this paper we perform a review of the state of the art methods used in multispectral facial recognition using images from the visible spectral band and also from the Near Infrared, Short Wavelength Infrared and Long Wavelength Infrared sub-bands. The public multispectral databases for facial analysis are identified, and a comparison is made, taking into consideration their specifications. The multispectral facial recognition methods are classified according to their basic working principle, from the traditional Fusion and Subspace methods to the more recent Deep Neural Networks.

Facial recognition is a method of identifying or authenticating the identity of people through their faces. Nowadays, facial recognition systems that use multispectral images achieve better results than those that use only visible spectral band images. In this work, a novel architecture for facial recognition that uses multiple deep convolutional neural networks and multispectral images is proposed. A domain-specific transfer-learning methodology applied to a deep neural network pre-trained in RGB images is shown to generalize well to the multispectral domain. We also propose a skin detector module for forgery detection. Several experiments were planned to assess the performance of our methods. First, we evaluate the performance of the forgery detection module using face masks and coverings of different materials. A second study was carried out with the objective of tuning the parameters of our domain-specific transfer-learning methodology, in particular which layers of the pre-trained network should be retrained to obtain good adaptation to multispectral images. A third study was conducted to evaluate the performance of support vector machines (SVM) and k-nearest neighbor classifiers using the embeddings obtained from the trained neural network. Finally, we compare the proposed method with other state-of-the-art approaches. The experimental results show performance improvements in the Tufts and CASIA NIR-VIS 2.0 multispectral databases, with a rank-1 score of 99.7% and 99.8%, respectively.