Advances in Super Resolution Reconstruction of Noisy, Blurred and Aliased Low Resolution Images

High Resolution images can be reconstructed from several blurred, noisy and aliased low resolution images using a computational process know as super resolution reconstruction. The ever increasing demand for more pixels or higher resolution in combination with the availability of more and more computational power has generated a large interest in super resolution imaging. The goal in super resolution imaging is to take multiple low resolution images of the same scene and combine them to generate a higher resolution image. During the process of our research we came across various super resolution algorithms; In this article we present the emergence and advances of super resolution techniques during the past 25 years. Existing survey papers [2],[3] on super resolution reconstruction provide an overview of the classical super resolution techniques where most of the aspects are ignored; in this review the recent advances in super resolution reconstruction and most of the theoretical and application settings of super resolution reconstruction often ignored in the past have been discussed. An outline of different super resolution techniques and two novel approaches for blind super resolution reconstruction using Discrete Cosine Transforms and Discrete Wavelet based Fusion and Interpolation are proposed. The proposed methods exhibits better super resolution factor compared to the traditional techniques. We also validate the results empirically and show that large enough magnification factor leads to overly smooth results.