A soft MAP framework for blind super-resolution image reconstruction

This paper proposes a new algorithm to address blind image super-resolution (SR) by fusing multiple low-resolution (LR) blurred images to render a high-resolution (HR) image. Conventional SR image reconstruction algorithms assume the blurring occurred during the image formation process to be either negligible or can be characterized fully a priori. This assumption, however, is impractical as it is often difficult to eliminate the blurring completely in some applications or to know the blurring function completely a priori. In view of this, we present a new soft maximum a posteriori (MAP) estimation framework to perform joint blur identification and HR image reconstruction. The proposed method incorporates a soft blur prior that estimates the relevance of the best-fit parametric blur model, and induces reinforcement learning towards it. An iterative scheme based on alternating minimization is developed to estimate the blur and the HR image progressively. Experimental results show that the new method is effective in performing blind SR image reconstruction where there is limited information about the blurring function.