本数据采用2022年哨兵2号光学卫星数据为数据源，按照区域及顺序，将全疆分为不同区域和不同类型图层。然后选择合适的深度学习模型依次对每一类要素制图，最后合并每种地类形成地表覆被。数据为CGCS2000坐标系，阿伯斯投影，精度为10米，共包括湖泊、森林、湿地、耕地、不透水层、冰雪、裸地、草地和其他等9个土地覆被类别。
", "ds_source": "采用2022年哨兵2号光学卫星数据为数据源。
", "ds_process_way": "采用2022年哨兵2号光学卫星数据为数据源，按照耕地、沙漠、冰川、湖泊、林地、草地、不透水层，裸地并按照区域及顺序，将全疆分为不同区域和不同类型图层。对每类要素，在全疆范围内选择样本，建立小样本库，通过样本增强、深度学习模型选择，得到全疆要素提取结果。按照该方法对每一地类要素进行深度学习提取，并将这一地类作为下一地类的掩膜，直至完成要求的所有地类。最后对信息提取的结果进行合并，得到完整的LUCC分类图。
", "ds_quality": "数据空间分辨率高，分类的空间细节丰富，类别清晰完整。不同要素的整体精度OA均在0.9以上，其中耕地、水体、冰川整体精度最高，草地次之，数据符合技术要求。
", "ds_acq_start_time": null, "ds_acq_end_time": null, "ds_acq_place": "", "ds_acq_lon_east": null, "ds_acq_lat_south": null, "ds_acq_lon_west": null, "ds_acq_lat_north": null, "ds_acq_alt_low": null, "ds_acq_alt_high": null, "ds_share_type": "login-access", "ds_total_size": 50856745256, "ds_files_count": 11, "ds_format": ".data格式", "ds_space_res": "10米", "ds_time_res": "", "ds_coordinate": "CGCS2000", "ds_projection": "Albers投影", "ds_thumbnail": "a228fb92-33c8-4b79-83eb-af8f19e0c7cd.png", "ds_thumb_from": 0, "ds_ref_way": "", "paper_ref_way": "", "ds_ref_instruction": "数据来源引用：新疆10m地表覆被数据集（2022年）来源于第三次新疆综合科学考察专项 \"空天地网一体化综合科考监测体系建设(2021xjkk1400)\"", "ds_from_station": null, "organization_id": "a5877b42-96ea-4f13-af7e-246f355413d6", "doi_value": "", "subject_codes": [ "170.45" ], "quality_level": 1, "publish_time": "2025-12-04 19:04:42", "first_publish_time": null, "last_updated": "2026-01-14 11:22:50", "protected": false, "protected_to": "2025-10-31 00:00:00", "lang": "zh", "cstr": "33110.11.XIEG.xjsedata.2022.00000105", "license": null, "extra": null, "files_shape": [ { "name": "2021xjkk1400-66-2023031066", "size": null, "is_dir": true } ], "features": null, "data_level": 0, "i18n": { "en": { "title": "Xinjiang 10m Land Cover Data Set (2022)", "ds_abstract": "This data uses the 2022 Sentinel 2 optical satellite data as the data source, and divides the entire Xinjiang into different areas and different types of layer layers according to regions and order. Then select the appropriate deep learning model to map each type of element in turn, and finally merge each land type to form the surface cover. The data is in the CGCS2000 coordinate system, Abers projection, with an accuracy of 10 meters, and includes 9 land cover categories including lakes, forests, wetlands, cultivated land, impervious layer, ice and snow, bare land, grassland and others.
", "ds_source": "Use 2022 Sentinel 2 optical satellite data as the data source.
", "ds_process_way": "Using the 2022 Sentinel 2 optical satellite data as the data source, the entire Xinjiang is divided into different areas and different types of layer layers according to cultivated land, desert, glacier, lake, forest land, grassland, impervious layer, bare land, and in accordance with the region and order. For each type of element, select samples within the entire Xinjiang, establish a small sample library, and obtain the extraction results of the entire Xinjiang element through sample enhancement and deep learning model selection. According to this method, deep learning and extraction are carried out on each ground class elements, and this ground class is used as a mask for the next ground class until all the required ground classes are completed. Finally, the results of information extraction are combined to obtain a complete LUCC classification map.
", "ds_quality": "The data spatial resolution is high, the classification is rich in spatial details, and the categories are clear and complete. The overall accuracy OA of different elements is above 0.9, among which cultivated land, water bodies and glaciers have the highest overall accuracy, followed by grassland. The data meets the technical requirements.
", "ds_ref_instruction": "Data source citation: Xinjiang's 10-meter surface cover dataset (2022) comes from the third Xinjiang comprehensive scientific expedition special project \"Construction of Integrated Comprehensive Scientific Research Monitoring System of Air, Space, Space and Network (2021 xjkk1400)\"", "ds_format": ".data format", "ds_projection": "Albers projection", "ds_space_res": "10 meters" } }, "license_type": null, "doi_reg_from": "reg_local", "cstr_reg_from": "reg_local", "doi_not_reg_reason": null, "cstr_not_reg_reason": null, "is_paper_in_submitting": false, "ds_topic_tags": [ "地表覆被" ], "ds_subject_tags": [ "地理学" ], "ds_class_tags": [], "ds_locus_tags": [ "新疆" ], "ds_time_tags": [ 2022 ], "ds_contributors": [ "李均力" ], "ds_meta_authors": [ "李均力" ], "ds_managers": [ "李均力", "李锦" ], "category": "土地利用/土地覆被" }