Classification of Fused Images using Radial Basis Function Neural Network for Human Face Recognition

摘要

Here an efficient fusion technique for automatic face recognition has beenpresented. Fusion of visual and thermal images has been done to take theadvantages of thermal images as well as visual images. By employing fusion anew image can be obtained, which provides the most detailed, reliable, anddiscriminating information. In this method fused images are generated usingvisual and thermal face images in the first step. In the second step, fusedimages are projected into eigenspace and finally classified using a radialbasis function neural network. In the experiments Object Tracking andClassification Beyond Visible Spectrum (OTCBVS) database benchmark for thermaland visual face images have been used. Experimental results show that theproposed approach performs well in recognizing unknown individuals with amaximum success rate of 96%.