<p>This dataset contains land use vector data from five Central Asian countries in 1990, 2000 and 2010, with a scale of 1: 100,000. The projection uses GCS_WGS1984. This data uses TM remote sensing data and Google earth data from 1989-1991, 1999-2000, and 2010 as data sources and verification data; using ARCGIS and Ecocognition software tools, TM remote sensing image data are preprocessed and automatically semi-automatically manually interpreted and classified, and verified by ground field surveys and Google earth high-precision data. The land cover classification is based on the "Central Asia Land Use/Cover Classification System", which refers to the Remote Sensing Land Classification System of the Institute of Geography and Resources, China Academy of Sciences (Liu Jiyuan) and the Ministry of Land and Resources initiated a project to study and formulate the "Land Use Status Classification" national standard. It adopts a two-level classification system: the first level is divided into 8 categories, mainly based on the utilization attributes of land resources, and the second level is divided into 30 types, mainly based on the operating characteristics, utilization methods and coverage characteristics of land resources. </p>
TM remote sensing data and Google earth data around 1989-1991, 1999-2000, and around 2010.
Using ARCGIS and ERDAS software tools, TM data corresponding to the age is selected, the data is preprocessed, and on the Ecocognition and Arcgis software platforms, the route of "segmentation-feature calculation-supervised classification-manual inspection and correction" is used for classification and interpretation. According to the "Remote Sensing Interpretation Manual for Comprehensive Scientific Expeditions in Central Asia"(Central Asia Land Use/Cover Classification System), remote sensing images were used for land use/cover classification and information extraction, and the extracted results were verified on the spot using field inspection data.
The land cover data is carefully and accurately evaluated through ground field surveys and high spatial resolution remote sensing data classification results. In order to ensure the validity of the evaluation results, in addition to accurate and reasonable reference data, the sample size, sampling method, sampling distribution and other effective sampling schemes are also designed. The ultimate goal of the scheme design is: Achieve an optimal balance between labor and equipment investment in image sampling and field sampling and the accuracy of analysis results, that is, obtain higher-precision analysis results with minimum economic investment. The entire operation process complies with relevant national standards, and the digital error is controlled within 2 pixels, ensuring that vector data can truly reflect the shape and attribute information of ground features, and the data accuracy is as high as 99%.
| # | number | name | type |
| 1 | 2010DFA92720 | Central Asia | International cooperation project of the Ministry of science and technology |
This work is licensed under a
Creative
Commons Attribution 4.0 International License.
CSTR:33110.11.ariddc.00259
10.12389/ariddc.00259
