本数据集基于深度学习的框架和开放获取数据(包括全球土地覆盖(GLC)产品、开放街道地图(OSM)和谷歌地球图像)建立了中国首个1米分辨率的国家尺度土地覆盖地图SinoLC-1。将三个10 m GLC产品和OSM数据结合生成可靠的训练标签。使用这些训练标签和来自Google Earth的1m分辨率图像来训练所提出的框架。该框架通过结合分辨率保持主干、弱监督模块和自监督损失函数，解决了图像和标签之间分辨率不匹配引起的标签噪声，从而在不需要人工标注的情况下自动改进VHR土地覆盖结果。基于大型存储和计算服务器，对73.25 TB数据集进行处理，获得覆盖全中国约960万平方公里的SinoLC-1。", "ds_source": "
数据来源于全球土地覆盖(GLC)产品、开放街道地图(OSM)和谷歌地球图像。", "ds_process_way": "
将三个10m GLC产品和OSM数据结合生成可靠的训练标签，使用这些训练标签和来自Google Earth的1m分辨率图像来训练所提出的框架。该框架通过结合分辨率保持主干、弱监督模块和自监督损失函数，解决了图像和标签之间分辨率不匹配引起的标签噪声，从而在不需要人工标注的情况下自动改进VHR土地覆盖结果。基于大型存储和计算服务器，对73.25TB数据集进行处理，获得覆盖全中国约960万平方公里的SinoLC-1。", "ds_quality": "
SinoLC-1产品使用包括超过10万个随机样本的视觉解释验证集和从中国政府提供的官方土地调查报告中收集的统计验证集进行验证。验证结果表明，SinoLC-1的总体准确率为73.61%，κ系数为0.6595。对各省区的验证进一步表明了该数据集在整个中国的准确性。此外，统计验证结果表明，SinoLC-1与官方调查报告一致，总体误估率为6.4%。此外，还将SinoLC-1与其他五种广泛使用的GLC产品进行了比较。结果表明，SinoLC-1具有最高的空间分辨率和最精细的景观细节。", "ds_acq_start_time": null, "ds_acq_end_time": null, "ds_acq_place": "中国", "ds_acq_lon_east": 136.44583333333333, "ds_acq_lat_south": 18.360555555555557, "ds_acq_lon_west": 70.00555555555556, "ds_acq_lat_north": 54.80416666666667, "ds_acq_alt_low": null, "ds_acq_alt_high": null, "ds_share_type": "login-access", "ds_total_size": 146062065323, "ds_files_count": 33, "ds_format": "tiff", "ds_space_res": "1m", "ds_time_res": "", "ds_coordinate": "无", "ds_projection": "", "ds_thumbnail": "643d320b-315a-4984-868b-be6aae30da55.png", "ds_thumb_from": 2, "ds_ref_way": "", "paper_ref_way": "", "ds_ref_instruction": "", "ds_from_station": null, "organization_id": "0a4269e1-65f4-45f1-aeba-88ea3068eebf", "ds_serv_man": "敏玉芳", "ds_serv_phone": "0931-4967596", "ds_serv_mail": "ncdc@lzb.ac.cn", "doi_value": "", "subject_codes": [ "170.45" ], "quality_level": 3, "publish_time": "2024-05-21 14:53:59", "last_updated": "2026-01-14 11:07:18", "protected": false, "protected_to": null, "lang": "zh", "cstr": "11738.11.NCDC.ZENODO.DB6468.2024", "i18n": { "en": { "title": "China's first 1-meter resolution national land cover dataset (SinoLC-1)", "ds_format": "tiff", "ds_source": "
The data is sourced from Global Land Cover (GLC) products, Open Street Maps (OSM), and Google Earth imagery.", "ds_quality": "
The SinoLC-1 product is validated using a visual interpretation validation set consisting of over 100000 random samples and a statistical validation set collected from official land survey reports provided by the Chinese government. The verification results indicate that the overall accuracy of SinoLC-1 is 73.61%, with a kappa coefficient of 0.6595. The validation of various provinces and regions further demonstrates the accuracy of this dataset throughout China. In addition, statistical verification results indicate that SinoLC-1 is consistent with the official survey report, with an overall misestimation rate of 6.4%. In addition, SinoLC-1 was compared with five other widely used GLC products. The results indicate that SinoLC-1 has the highest spatial resolution and the finest landscape details.", "ds_ref_way": "", "ds_abstract": "
This dataset is based on a deep learning framework and open access data, including Global Land Cover (GLC) products, Open Street Maps (OSM), and Google Earth imagery, to establish China's first 1-meter resolution national scale land cover map, SinoLC-1. Combine three 10 meter GLC products with OSM data to generate reliable training labels. Use these training labels and 1m resolution images from Google Earth to train the proposed framework. This framework solves the label noise caused by resolution mismatch between images and labels by combining resolution preserving backbone, weakly supervised module, and self supervised loss function, thereby automatically improving VHR land cover results without the need for manual annotation. Based on large-scale storage and computing servers, the 73.25 TB dataset was processed to obtain SinoLC-1 covering approximately 9.6 million square kilometers throughout China.
", "ds_time_res": "", "ds_acq_place": "China", "ds_space_res": "1m", "ds_projection": "", "ds_process_way": "Combine three 10m GLC products with OSM data to generate reliable training labels, and use these labels along with 1m resolution images from Google Earth to train the proposed framework. This framework solves the label noise caused by resolution mismatch between images and labels by combining resolution preserving backbone, weakly supervised module, and self supervised loss function, thereby automatically improving VHR land cover results without the need for manual annotation. Based on large-scale storage and computing servers, the 73.25TB dataset was processed to obtain SinoLC-1 covering approximately 9.6 million square kilometers throughout China.", "ds_ref_instruction": "" } }, "submit_center_id": "ncdc", "data_level": 0, "license_type": "https://creativecommons.org/licenses/by/4.0/", "doi_reg_from": "reg_outside", "cstr_reg_from": "reg_outside", "doi_not_reg_reason": null, "cstr_not_reg_reason": null, "is_paper_in_submitting": false, "ds_topic_tags": [ "1m分辨率", "土地覆盖", "深度学习" ], "ds_subject_tags": [ "地理学" ], "ds_class_tags": [], "ds_locus_tags": [ "中国" ], "ds_time_tags": [], "ds_contributors": [ { "true_name": "李卓鸿", "email": "hunter.lee@whu.edu.cn", "work_for": "武汉大学测绘遥感信息工程国家重点实验室", "country": "中国" }, { "true_name": "张洪艳", "email": "zhanghongyan@cug.edu.cn", "work_for": "中国地质大学（武汉）计算机学院", "country": "中国" } ], "ds_meta_authors": [ { "true_name": "李卓鸿", "email": "hunter.lee@whu.edu.cn", "work_for": "武汉大学测绘遥感信息工程国家重点实验室", "country": "中国" } ], "ds_managers": [ { "true_name": "张洪艳", "email": "zhanghongyan@cug.edu.cn", "work_for": "中国地质大学（武汉）计算机学院", "country": "中国" } ], "category": "水文" }