A Novel Super-resolution Reconstruction Algorithm Based on Subspace Projection

Increasing image resolution is a challenging and fundamental image-fusion operation, super-resolution a technique to increase image resolution. However, most of method can not reconstruct image from un-regular sampled data. In this paper, we propose a novel super-resolution reconstruction algorithm based on polynomial bases. This algorithm can be combined with intensity information and structural details in the image. The density of sampled data and local structure decide the certainty function and structural-adaptive applicability function of neighbourhood. Experimental results show that the proposed algorithm can improve denoising effect in the super-resolution reconstruction image, and achieve a state-of-the-art high-resolution visual effect in the edges and detailed features of image.