Generalized interpolation and its application in super-resolution imaging

Abstract

In this paper, we present a generalized interpolation scheme for image expansion and generation of super-resolution images. This is done by decomposing the image into appropriate subspaces, carrying out interpolation in individual subspaces and subsequently transforming the interpolated values back to the image domain. Various optical and structural properties of the image, such as 3-D shape of an object, regional homogeneity, local variations in scene reflectivity, etc., can be better preserved during the interpolation process. The motivation for doing so has also been explained theoretically. The generalized interpolation scheme is also shown to be useful in perceptually based high-resolution representation of images where interpolation is done on individual groups as per the perceptual necessity. Further, this scheme is also applied to generation of high-resolution transparencies from low resolution transparencies.