Regularization based super-resolution image processing algorithm using edge-adaptive non-local means filter
DaeOk Kim, Hyeran Byun · 2013
Super-Resolution (SR) image reconstruction is a technology to reconstruct multiple low-resolution images into one or multiple high-resolution images. As the use of digital camera is recently increasing, the advancement of super-resolution technology gets a great attention. In this study, we propose a regularization-based Super-Resolution algorithm that utilizes an Edge-adaptive Non-Local Means filter. We compare the result of image reconstruction through the algorithm that we proposed and that of image reconstruction through existing studies. As a result, we could verify that a better result would be obtained for regularization function when using an Edge-adaptive Non-Local Means filter rather than using a Non-Local Means filter. We could also obtain much higher PSNR(Peak Signal-to Noise Ratio) than using a Bilateral Total Variation(BTV) method.