Edge-directed interpolation-based sub-pixel mapping

Sub-pixel mapping is a technology to obtain the spatial distribution information of land cover within mixed pixels. In this letter, two kinds of edge-directed interpolation algorithms are proposed. These algorithms are applied to coarse spatial resolution images to obtain a high-resolution image with probability information. Then hard classification on a per sub-pixel basis is implemented to achieve sub-pixel mapping. Both methods are demonstrated with three synthetic images. Experiments show that the proposed sub-pixel mapping methods can retain image edge feature better and produce higher accuracy.