International Journal of Emerging Science and Engineering (IJESE)
In this paper, a face recognition method based on simultaneous sparse approximations under varying illumination is used. This method consists of two main stages. In the first stage, a dictionary is learned for each face class based on given training examples which minimizes the representation error with a sparseness constraint. In the second stage, a novel image is projected onto the span of the atoms in each learned dictionary. The resulting residual vectors are then used for classification. Furthermore to handle variations in lighting conditions an image relighting technique based on a non-stationary stochastic filter is used to generate multiple frontal images of the same person with variable lighting.