Zoom-based super-resolution image reconstruction from images with different orientations

Abstract

Construction of a mosaic image using a set of low resolution images taken at different zoom settings and at different angles was investigated in this thesis. The proposed reconstruction algorithm uses the zoom based super resolution technique, based on maximum likelihood estimate. A computationally less intensive point matching algorithm was introduced based on a known algorithm. The simulation results show that the implemented algorithm can find point correspondences of images successfully, even they are differently zoomed, that helps for image registration.