Further improvement of super-resolution reconstruction

We propose a novel approach to improve further the quality of recovered images from standard super-resolution reconstruction, using Lewitt's Kaiser-Bessel window functions (blobs) as the basis functions instead of normal pixels or voxels. The spatially localised and rotationally symmetric properties of blobs have made them very attractive for iterative image reconstruction. However, these same properties of blobs can be also very advantageous for super-resolution recovery, when more than one of a similar 2D or 3D scene is available. We show in this paper that by incorporating blobs into the super-resolution algorithm for image recovery; we can obtain much better quality, especially when there are only a few lower quality images available for the same scene. Moreover, using fewer low-resolution images for super-resolution reconstruction, we can also guarantee improvement in computational time.