本数据集面向高海拔寒区公路在多年冻融与强环境扰动条件下的病害识别、机理解析与风险评估需求，基于2019-2023年公路病害实地调研数据，围绕青藏公路（格尔木—安多）、新藏公路（叶城—拉孜）与川藏公路北线（那曲—昌都）建立2019—2023年本底数据集，通过统一坐标与路网里程参考、质量控制与时空配准，形成2年时间分辨率、1 km空间分辨率的连续化路域本底表征；其次在典型病害路段尺度上，选取15个代表性路段构建衍生数据集，基于高分辨率影像/点云等精细观测与标准化标注流程，提取病害类型、几何形态与空间分布等关键衍生指标，形成2年时间分辨率、10 cm空间分辨率的病害精细表征产品。数据内容包括：宏观路域层面的多年度本底信息，以及典型路段层面的病害类型与几何特征等衍生信息，可用于多尺度病害时空演化分析、模型训练与泛化验证，以及高寒公路智能巡检与养护决策支持，为高海拔地区公路韧性提升与精细化养护提供数据基础。", "ds_source": "
本数据集以现场病害调查为核心数据源，按2年周期开展重复调查；调查过程中结合无人机/人工巡检获取路面近景影像与定位信息，并以道路里程为基准进行路段分段记录；随后依据统一的病害分类标准对裂缝、坑槽、沉陷等进行判读标注，提取其类型与发育状况，并通过多期对齐与一致性校核形成可比的时序化调查数据。", "ds_process_way": "
对实地调查数据进行坐标和里程统一，按路段编号整理，剔除重复与异常，补齐缺失；再按2年时间步将信息汇总到1 km路域单元；汇总病害类型及长度、宽度、面积等几何指标；最后进行多期对齐与复核，形成结构化数据表与配套元数据。", "ds_quality": "
数据比较准确，达到精度要求。", "ds_acq_start_time": "2019-06-15 00:00:00", "ds_acq_end_time": "2023-08-23 00:00:00", "ds_acq_place": "青藏公路（格尔木—安多）,新藏公路（叶城—拉孜）,川藏公路北线（那曲—昌都）", "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": 4752795, "ds_files_count": 8, "ds_format": "excel", "ds_space_res": "1km", "ds_time_res": "2年", "ds_coordinate": "无", "ds_projection": "", "ds_thumbnail": "bc991fa1-b98d-4983-81f2-9cff42f8cbc2.png", "ds_thumb_from": 2, "ds_ref_way": "", "paper_ref_way": "", "ds_ref_instruction": "", "ds_from_station": null, "organization_id": "76330c66-832b-46b3-b501-f5f6edb08dc2", "ds_serv_man": "敏玉芳", "ds_serv_phone": "0931-4967596", "ds_serv_mail": "ncdc@lzb.ac.cn", "doi_value": "", "subject_codes": [ "410" ], "quality_level": 3, "publish_time": "2026-01-28 18:08:37", "last_updated": "2026-02-05 17:03:08", "protected": false, "protected_to": null, "lang": "zh", "cstr": "11738.11.NCDC.NIEER.DB7064.2026", "i18n": { "en": { "title": "Highway Infrastructure Damage Dataset for High-Altitude Regions (2019–2023)", "ds_format": "excel", "ds_source": "
This dataset is based on on-site disease investigation as the core data source, with repeated surveys conducted every 2 years; During the investigation process, close range images and positioning information of the road surface were obtained by combining unmanned aerial vehicles/manual inspections, and segmented records were made based on the road mileage; Subsequently, based on a unified disease classification standard, cracks, pits, subsidence, etc. were interpreted and annotated, and their types and development status were extracted. Comparable time-series survey data were formed through multi period alignment and consistency verification.", "ds_quality": "
The data is relatively accurate and meets the accuracy requirements.", "ds_ref_way": "", "ds_abstract": "
This dataset is aimed at the needs of disease identification, mechanism analysis, and risk assessment of high-altitude cold region highways under years of freeze-thaw and strong environmental disturbance conditions. Based on the field survey data of highway diseases from 2019 to 2023, a background dataset from 2019 to 2023 is established around the Qinghai Tibet Highway (Golmud Anduo), Xinzang Highway (Yecheng Lazi), and the northern Sichuan Tibet Highway (Nagqu Changdu). Through unified coordinates and road network mileage reference, quality control, and spatiotemporal registration, a continuous road domain background representation with 2-year time resolution and 1 km spatial resolution is formed; Secondly, at the scale of typical disease prone road sections, 15 representative road sections were selected to construct a derived dataset. Based on high-resolution images/point clouds and standardized annotation processes, key derived indicators such as disease types, geometric shapes, and spatial distributions were extracted to form a 2-year time resolution and 10 cm spatial resolution disease fine characterization product. The data content includes: multi-year background information at the macro road domain level, as well as derived information such as disease types and geometric features at the typical road segment level. It can be used for multi-scale disease spatiotemporal evolution analysis, model training and generalization verification, as well as intelligent inspection and maintenance decision support for high-altitude highways, providing a data foundation for improving the resilience and fine maintenance of highways in high-altitude areas.", "ds_time_res": "2年", "ds_acq_place": "Qinghai Tibet Highway (Golmud Anduo), Xinjiang Tibet Highway (Yecheng Lazi), North Sichuan Tibet Highway (Nagqu Changdu)", "ds_space_res": "1km", "ds_projection": "", "ds_process_way": "
Unify the coordinates and mileage of field survey data, organize them by road section number, eliminate duplicates and anomalies, and fill in missing information; Summarize the information into a 1 km road domain unit in a 2-year time step; Summarize the types of diseases and geometric indicators such as length, width, and area; Finally, perform multi period alignment and review to form structured data tables and supporting metadata.", "ds_ref_instruction": "" } }, "submit_center_id": "ncdc", "data_level": 0, "license_type": "CC BY 4.0", "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": [ "路面病害", "病害调查数据", "高海拔公路", "多期时序" ], "ds_subject_tags": [ "工程与技术科学基础学科" ], "ds_class_tags": [], "ds_locus_tags": [ "青藏公路", "新藏公路", "川藏公路北线" ], "ds_time_tags": [ 2019, 2020, 2021, 2022, 2023 ], "ds_contributors": [ { "true_name": "柴明堂", "email": "cmt620422@163.com", "work_for": "宁夏大学", "country": "中国" } ], "ds_meta_authors": [ { "true_name": "柴明堂", "email": "cmt620422@163.com", "work_for": "宁夏大学", "country": "中国" } ], "ds_managers": [ { "true_name": "柴明堂", "email": "cmt620422@163.com", "work_for": "宁夏大学", "country": "中国" } ], "category": "其他" }