Remote Sensing Image Segmentation of Ulan Buh Desert Based on Mathematical Morphology

As the basis of 1:10000 color aerial remote sensing image of Ulan Buh Desert, Inner Mongolia, this paper studies on white thorn which is a typical representative of the nature of scarcity and mass organizations like the vegetation in the desert region. By using mathematical morphology theory and the Matlab 7.0 software, the white thorn will be conducted from the remote sensing image segmentation. Comparing the conducted image with the outer shape of the industry survey results, the accuracy of segmentation could be verified by the gray-level co-occurrence matrix comparisons. This research provides the technical reference for the large sample survey of desert region by remote sensing image segmentation.