The International Journal of Innovative Research in Computer and Communication Engineering
In this paper, the authors propose a novel method for recognizing faces degraded by blur and illumination variation using set-theoretic characterization method. In the unconstrained environment, face recognition becomes difficult because of Blur, noise and illumination. In this paper, they address problem of face recognition in unconstrained domain. The convex set is set of all images which are obtained by blurring a given image. In this algorithm, various blur characteristics are taken under consideration. By blurring image by various blur function convex set is formed and that corresponding convex set is associated with each gallery image.