Multi-frame Image Super-Resolution by Interpolation and Iterative Backward Projection

Abstract

Image Super-Resolution is the well-known technique of improving the quality of images by sharpening the resolution of images. Multi- frame Super- Resolution is procedure of developing an image with high resolution by combining a series of poor resolution images of the look-alike scene. Here, an Improved Super- Resolution technique is implemented using interpolation and iterative backward projection (IBP) registration algorithm. Initially, bilinear interpolation is carried out for a sequence of four images with poor resolution. These resultant images are then registered using Iterative Backward Projection algorithm to yield a high-resolution image. This Improved Super- resolution reconstruction algorithm is simulated on MATLAB and compared with Interpolation Reconstruction algorithm. Improved Super Resolution technique proves to be excellent than Interpolation Reconstruction techniques by means of visuality, Mean_square_error (MSE), Peak-Signal to Noise ratio (PSNR).