Single Image Super Resolution Method Based on Edge Preservation

In this paper, we present a novel super resolution (SR) framework based on edge preservation. The iterative back-projection algorithm (IBP) is a classical SR method and has low computational complexity, which can be applied in real time applications. However, it often produces many artifacts especially along the strong edges. To reduce the jaggy artifacts, our approach has three steps. First, we improve the initial estimate using bilateral filtering to strengthen the true edges. Second, we learn the structural content of low resolution pixel and the correlation among the pixels with similar structure. Third, we use the correlation to guide the output image reconstructed by the IBP algorithm. The experimental result proved that our proposed method can remove the artifacts and obtain clear and sharp edges in visual perception.