Investigation into Standard Image Restoration Methods and their Comparison with Proposed Technique for Deblurring Images

The restoration of digital image is very important process in the image processing.. Image restoration attempts to restore images that have degraded because of camera shake, movement of object or both. The degradation mainly involves blurring and noise. By combining information extracted from both blurred and noisy images we show in this paper how to produce a high quality image that cannot be obtained by simply denoising the noisy image, or deblurring the blurred image alone. To remove blurring, we need to (i) know and analyse how the image is blurred (ii) restore a natural looking image through deconvolution formulas. In deconvulation the main thing is blur kernel or Point Spread Function. Estimating Blur kernel is really very challenging because the algorithm needs to distinguish between correct image pairs and Incorrect ones. Difficulty also arises as the because the algorithm needs to restore high frequency image contents attenuated by blur. In this paper, we shall deal with a few of these challenges. We will introduce an technique that shows blur kernel can be estimated by analyzing blurred edges i-e we shall analyze blurred edges to calculate Radon Transform and then recover the blur kernel using the inverse of Radon transform. This method is effective and is well suited to images with many edges. We introduce a method to integrate this data into a maximum-a-posteriori kernel estimation framework and show its pros and cons. In this paper we compare Restored Blurred Images by using four types of techniques of deblurring images such as Wiener filter, Regularised filter, Lucy Richardson deconvlution algorithm and our proposed algorithm on the basis of Wellknown image quality accessment parameters like mean squared error (MSE) and peak signal-to-noise ratio (PSNR).