Improved motion-based localized super resolution technique using discrete wavelet transform for low resolution video enhancement

Super resolution is used for resolution enhancement of images or video sequences. Instead of super resolving frames globally, using localized motion based super resolution increases the quality of the enhanced frames. In this paper, we propose to use the super resolution on different subbands of localized moving regions extracted from discrete wavelet transform (DWT) and composing the super resolved subbands using inverse DWT (IDWT) to generate the respective enhanced high resolution frame. The results based on PSNR values, in comparison with the global super resolution method, show improvement in quality. The improvement is achieved by isolating different frequencies in different subbands extracted from DWT and super resolving them separately.