An Integrated Land Classification Approach for the Giant Panda National Park

National park is an important material basis for the construction of ecological civilization, and the land feature covering situation of national park will be a significant reference data. However, the land classification approach mainly focused on the algorithm research and utilised the manual verification, which will be time-consuming and difficult to extract and verify the land feature information in the experimental region of Giant panda national park. Therefore, a new land classification approach by integrating the remote sensing orthorecified data, 1:10000 basic geographic information data and monitoring data of geographical conditions, which can successfully extract the land features in the Giant panda national park, was proposed in this paper. This proposed method showed that the coverage of woodland in the experimental region was highest, and the acreage of woodland accounted for 98.05% of total acreage, following by other lands, grassland, surface water, garden and cultivated land, and the evaluation conclusion of land classification ratio was "qualified". Because the experimental areas of Giant panda national park selected in this paper was large, the proposed integrated land classification approach recommended in this paper can be used as a significant method when conducting the land classification for large acreage of national park.