Abstract

This research designs and implements a tool that allows the combination of state of the art super-resolution algorithms in a flexible way to facilitate the search for high quality single frame super-resolution algorithms. The ability to combine algorithmic parts in a flexible way decreases the time needed to design new super-resolution algorithms. Current state of the art super-resolution algorithms are researched and broken down into elemental parts. A modular approach is designed guided by these elemental parts and a proof of concept toolkit is implemented to perform testing. It offers the ability to design super-resolution algorithms without requiring any programming or scripting. Promising test results of a new super-resolution approach based on directional information are included.