This data set contains the oasis data of Sangong River Basin in 1970s, 1990s, 2000 and 2010. After the establishment of the interpretation mark, the oasis information can be extracted for image interpretation. In the process of interpretation, cbres image, land cover, auxiliary data and corresponding topographic map in corresponding period should be taken as the comprehensive information source, and the interpretation and analysis should be completed by professionals. In the attribute judgment of land cover type, the accuracy rate of water body, farmland and construction land should reach or exceed 90%, and that of other types should reach or exceed 80%. When drawing the type boundary, the error between the remote sensing image and the main types of land cover is generally less than 0.6 mm on the map, and the maximum error is no more than 1 mm on the map. Data file name: 1. Sangonghe Oasis (1) SG_ oasis_ 1970 （2）sg_ oasis_ 1990 (3）sg_ oasis_ 2000 （4）sg_ oasis_ two thousand and ten
USGS data acquisition
1. The remote sensing data are LANDSAT MSS and TM / ETM data. The corrected multispectral images are synthesized by 432 band pseudo color. The oasis information is extracted by 1:100000 mapping standard in ArcGIS software by human-computer interaction. 2. Data generation process: image download geometric correction classification system determination interpretation mark determination human computer interaction interpretation accuracy verification
95 patch attributes were wrong, and the qualitative accuracy of interpretation was 97.44%. The accuracy of each type is more than 90%. The accuracy of water wetland and construction land was 100% and 98.51%, the accuracy of cultivated land and woodland was 96.85% and 96.29%. The accuracy of grassland was 93.81% because of the low coverage grassland.
| # | number | name | type |
| 1 | 2009CB421301 | National Program on Key Basic Research Project (973 Program) | |
| 2 | 2009CB421300 | National Program on Key Basic Research Project (973 Program) |
This work is licensed under
CC BY 4.0 (Creative Commons Attribution 4.0 International License).
CSTR:11738.11.ncdc.973Desert.2020.161
10.12072/ncdc.973Desert.db0008.2020
