A New Approach for Retrieving Unconstrained Blurred Images?

Face Recognition has been of great importance in different fields of technology. But when blurring come to picture face recognition becomes difficult. There has many approaches proposed in this context like blind image deconvolution, image statistics etc. But all these approaches concentrate mainly on the blurring part alone. Here in this approach the illumination defects are together considered with the blur problems. With the calculation of minimum distance from the given blurred image to artificially blurred image the corresponding image can be mapped. Incorporation of the image features under different illumination conditions makes the face recognition much easier. To improve the performance, characteristics of the blur can be added along with. Weights are assigned to different pixels based on the priority. L1 norm is followed as it is robust misalignments in the pixels. Finally it can be seen that the method is not that complex even though it incorporates both blur and illumination but gives more clarity and perfection.