DEBLURRING OF VIDEO SEQUENCES

Abstract

The development of visual information for human logic and processing of visual data for independent machine perception are the main important areas that had triggered the interest in image processing subject long ago. Image restoration process refers to elimination or minimization of degradations in an image. The process of image restoration includes deblurring of images which are degraded by the various limitations of camera sensor, environment, correction of geometric distortion or non-linearity’s and noise due to camera sensors. Image deblurring/deconvolution is a process of recovering the original image from a degraded image. Point spread function(PSF) is the function which causes the degradation of images also called as the blur kernel or degradation function. The light or electromagnetic waves that we intend to capture propagates from a point source. The light spreads due to environmental or imaging system reasons thus reducing the quality of imaging system. This problem is ill conditioned by nature. We propose a method where we use maximum a posterior approach and also use the information in the spectrum of the image to deblur the image. The proposed method is robust and is able to handle a variety of blurs.