Classification of Polar-Thermal Eigenfaces using Multilayer Perceptron for Human Face Recognition

This paper presents a novel approach to handle the challenges of face\nrecognition. In this work thermal face images are considered, which minimizes\nthe affect of illumination changes and occlusion due to moustache, beards,\nadornments etc. The proposed approach registers the training and testing\nthermal face images in polar coordinate, which is capable to handle\ncomplicacies introduced by scaling and rotation. Polar images are projected\ninto eigenspace and finally classified using a multi-layer perceptron. In the\nexperiments we have used Object Tracking and Classification Beyond Visible\nSpectrum (OTCBVS) database benchmark thermal face images. Experimental results\nshow that the proposed approach significantly improves the verification and\nidentification performance and the success rate is 97.05%.\n