延安地区无人机探地雷达探测科学数据，采用无人机搭载100MHz探地雷达进行黄土滑坡地下结构非破坏性探测，用于识别黄土滑坡地下落水洞、裂隙等结构。共进行了两项试飞实验，实验1为大范围快速探测实验，试验2为中等范围探测实验。针对飞行过程中采集到的原始数据设计了包含背景去除、增益、带通滤波、中值滤波、自适应滤波的数据处理流程。对比处理前后图像可见，发射和接收天线间的耦合得到了较好的压制，来自地表之下的信号能量得到了较好的补偿，原始数据中的随机噪声干扰也都得到了压制。
", "ds_source": "数据来源于延安市安塞区湫滩，采用无人机搭载100MHz探地雷达进行黄土滑坡地下结构非破坏性探测，用于识别黄土滑坡地下落水洞、裂隙等结构，飞行高度控制在5-15米，飞行速度保持在2米/秒，全过程严格遵守设备维护和操作规范，所有数据均按标准格式存档备份，确保数据的完整性、准确性和可追溯性，有效支撑后续地下结构分析和科研应用。
", "ds_process_way": "通过软件进行背景去除、增益、带通滤波、中值滤波、自适应滤波的数据处理，从而完成实际图像的呈现。
", "ds_quality": "为了定量确定无人机探地雷达的有效探测深度，我们引入了Xing 等人2017年提出的探地雷达穿透深度估计方法。要估算穿透深度，需要用一个滑动窗口计算每两个相邻轨迹的互相关性。互相关结果变化最快的时间位置则表示穿透深度。我们通过计算所有道互相关性的局部导数来表示变化的快慢，
\n验证结果表明，表面地下介质较为均匀，穿透深度曲线较好地刻画了有信号区域与无信号区域之间的边界，无人机探地雷达能探测黄土体地下5m深度的结构，最深处可达8m。
", "ds_acq_start_time": "2024-08-12 00:00:00", "ds_acq_end_time": "2024-08-13 00:00:00", "ds_acq_place": "延安市安塞区", "ds_acq_lon_east": 109.43833333333333, "ds_acq_lat_south": 36.5125, "ds_acq_lon_west": 108.86222222222221, "ds_acq_lat_north": 37.32527777777778, "ds_acq_alt_low": null, "ds_acq_alt_high": null, "ds_share_type": "apply-access", "ds_total_size": 430921894, "ds_files_count": 0, "ds_format": "*.zry", "ds_space_res": "", "ds_time_res": "", "ds_coordinate": "无", "ds_projection": "", "ds_thumbnail": "501ee6d4-d712-4dac-aa73-d93907baa861.png", "ds_thumb_from": 0, "ds_ref_way": "", "paper_ref_way": "", "ds_ref_instruction": "", "ds_from_station": "", "organization_id": "bf138922-7121-438c-8d1b-19d5f751c907", "ds_serv_man": "李红星", "ds_serv_phone": "0931-4967592", "ds_serv_mail": "ncdc@lzb.ac.cn", "doi_value": "", "subject_codes": [ "170.2540" ], "quality_level": 0, "publish_time": "2026-05-21 17:29:59", "last_updated": "2026-05-21 17:35:11", "protected": false, "protected_to": null, "lang": "zh", "cstr": "11738.11.ncdc.loess.db7344.2026", "i18n": { "en": { "title": "Scientific Data from Drone Ground-Penetrating Radar Detection in Yan'an Region", "ds_format": "*.zry", "ds_source": "The data were collected from the Qiu Tan area in Ansai District, Yan'an City. Using a drone equipped with a 100 MHz ground-penetrating radar (GPR), non‑destructive subsurface detection of loess landslides was conducted to identify underground features such as sinkholes and fractures. The flight altitude was maintained between 5 and 15 meters, and the flight speed was kept at 2 m/s. Throughout the process, equipment maintenance and operational protocols were strictly followed. All data were archived and backed up in standard formats to ensure integrity, accuracy, and traceability, thereby effectively supporting subsequent subsurface structural analysis and scientific research applications.", "ds_quality": "
In order to quantitatively determine the effective detection depth of the drone-based ground-penetrating radar (GPR), we introduced the GPR penetration depth estimation method proposed by Xing et al. in 2017. To estimate the penetration depth, a sliding window was used to compute the cross-correlation between each pair of adjacent traces. The time position at which the cross-correlation result changed most rapidly indicates the penetration depth. We quantified the rate of change by calculating the local derivative of the cross-correlation of all traces. The verification results show that the subsurface medium near the surface is relatively uniform, and the penetration depth curve effectively delineates the boundary between signal-present and signal-absent zones. The drone-based GPR is capable of detecting structures up to 5 m deep in loess, with the maximum depth reaching 8 m.", "ds_ref_way": "", "ds_abstract": "
This study employed a drone-mounted 100 MHz ground-penetrating radar (GPR) to conduct non‑destructive detection of the subsurface structure of loess landslides, aiming to identify underground features such as sinkholes and fractures. Two test‑flight experiments were carried out: Experiment 1 involved large‑area rapid detection, while Experiment 2 focused on medium‑range detection. A data‑processing workflow was designed for the raw data collected during the flights, which included background removal, gain adjustment, band‑pass filtering, median filtering, and adaptive filtering. Comparison of the images before and after processing shows that the coupling between the transmitting and receiving antennas was effectively suppressed, the signal energy from below the ground surface was well compensated, and random noise interference in the raw data was also significantly reduced.", "ds_time_res": "", "ds_acq_place": "Ansai District, Yan'an City", "ds_space_res": "", "ds_projection": "", "ds_process_way": "
Software-based data processing was conducted through background removal, gain adjustment, band‑pass filtering, median filtering, and adaptive filtering, thereby achieving the visualization of practical images.", "ds_ref_instruction": "" } }, "submit_center_id": "ncdc", "data_level": 0, "recommendation_value": 0, "license_type": "https://creativecommons.org/licenses/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": [ "探地雷达", "无人机", "matlab" ], "ds_subject_tags": [ "空间物理探测" ], "ds_class_tags": [], "ds_locus_tags": [ "延安市安塞区" ], "ds_time_tags": [ 2024 ], "ds_contributors": [ { "true_name": "冯晅", "email": "fengxuan@jlu.edu.cn", "work_for": "吉林大学", "country": "中国" } ], "ds_meta_authors": [ { "true_name": "冯晅", "email": "fengxuan@jlu.edu.cn", "work_for": "吉林大学", "country": "中国" } ], "ds_managers": [ { "true_name": "冯晅", "email": "fengxuan@jlu.edu.cn", "work_for": "吉林大学", "country": "中国" } ], "category": "其他" }