Image Super Resolution Technique based on Adaptive Lifting and Stationary Wavelet Decomposition

摘要

A novel algorithm for image superresolution based on the interpolation of high frequency subbands with adaptive lifting wavelet transform (ALWT) and stationary wavelet transform (SWT) is introduced. ALWT is used in order to decompose an input image into different subbands. SWT is introduced to correct the estimated coefficients. The edges are enhanced by the increment of the high frequency subbands obtained by ALWT and SWT. Afterwards, all these images are combined using ILWT to generate a super resolved image. In order to evaluate the image quality, MSE and PSNR are both computed for the resolved image. Quantitative and qualitative experiments for the proposed method showed great superiority in obtaining high resolution images with high PSNR and much better visual quality.