Image Deblurring in Bayesian Framework Using Template Based Blur Estimation
Image deblurring aims to reconstruct a high quality image from a degraded image. For deblurring, the blurring function which causes the degradation is estimated first. A template based blur estimation method is presented here which can identify Gaussian and Uniform blur. Due to the ill-posed nature of Image Restoration (IR) process, prior knowledge of natural images is used to regularize the IR problem. The Bayesian approach provides the means to incorporate prior knowledge in data analysis. A comparative analysis using various priors was studied qualitatively. PSNR, SNR, SSIM and IQI are the performance measures used.