A dataset of land cover classifications with a spatial resolution of 30m in Mongolia in 2005 and 2015

Based on the acquired Landsat remote sensing image, ENVI5.2 software is used for radiation correction and geometric correction of the original image to eliminate image distortion and deformation. Using eCognition software, multiscale segmentation and spectral difference segmentation of remote sensing images are carried out. Then, make full use of the relationship between the spectrum, texture and geographical environment of the ground objects, calculate the normalized vegetation index (NDVI), normalized water index (NDWI) and soil brightness index (NDWI), and classify the ground objects according to specific rules to obtain the single scene remote sensing image interpretation results. After classification and post-processing of single scene image interpretation results, Mongolia land cover classification data products are obtained. This data set contains the land cover classification data of Mongolia with a resolution of 30m in 2005 and 2015. The data format is img format, and the geographical range is the whole territory of Mongolia (87 ° 44 '- 119 ° 56'E, 41 ° 35' - 52 ° 09'N).