A Hyper-Resolution Scheme for Enhancement on a Down-Sampled Image

The quality of an image can be evaluated on its spatial and spatial-frequency resolutions. Image interpolation and super-resolution are perhaps the way to respectively produce high spatial and spatialfrequency resolutions of an image, especially for a down-sampled image. In this paper, we first address an approach to interpolating an image with noise removal by exploiting the probability filter coupled with a pyramidal decomposition. Then,the high spatial-frequency spectrum of any interpolated image can be reconstructed using the improved Poisson maximum a posteriori (MAP) superresolution algorithm. This hybrid scheme called “hyper-resolution” here is proposed for enhancing the spatial and spatial-frequency resolutions of down-sampled images. Utilizing the hyperresolution scheme, this research has studied resolution enhancement of gray-level and color images. Simulation results demonstrate that this hyper-resolution scheme substantially improves the quality of down-sampled images.