Spatially adaptive superresolution using the optimal recovery framework

Multi-Frame image restoration is a form of Super-Resolution (SR) which consists of combining multiple Low-Resolution (LR) images in order to reconstruct a single High-Resolution (HR) image. Our previous approach to SR based on the framework of optimal recovery is extended to a spatially adaptive scheme whereby the block-by-block processing is modified based on the properties of the corresponding local LR image data. The bandwidth parameter is adapted based on the local variance of each corresponding block from all LR images that contribute to the HR image block to be reconstructed. Also, an optimal regularization parameter for the reconstruction of each HR output block is calculated using the Generalized Cross Validation (GCV) approach. This parameter is dependent on the chosen bandwidth as well as the contributing local LR pixels values. Simulation results show the superiority of the adaptive scheme over the fixed optimal recovery based super-resolution with and without the use of overlapping blocks. A brief comparison with other SR schemes is also included to illustrate the possible performance.