Marine Target Data Sharing Program(MTDSP)——Multisource Observation Dataset of Marine Targets

Radar sea detection data - The multi-source observation test dataset for sea targets covers 7 typical types of sea targets, including dry cargo ships, liquid cargo ships, container ships, passenger ships, fishing boats, engineering ships, and search and rescue ships, with 4 groups per type, totaling 28 sets of data (ZIP files), with a total data volume of approximately 42GB. The data content includes multi-source observation information, including radar HH/VV polarized intermediate frequency/video echoes, visible light and infrared images, target trajectories (radar and AIS), and meteorological and hydrological parameters (wind speed, wind direction, wave height, wave direction), covering different sea conditions, weather conditions, and target movement states. It has high spatiotemporal resolution and data integrity, providing high-quality data support for maritime target detection, recognition, and motion analysis research. The data collection equipment is installed at the first beach in Yantai. The radar adopts a circular scanning mode for long-term continuous observation, and dynamically intercepts local area data based on real-time target position information to optimize storage and parsing efficiency and reduce redundant data burden. The optoelectronic device is in passive mode, receiving real-time target distance, azimuth, and elevation information provided by the radar, and automatically adjusting its own parameters to ensure accurate alignment of multi-source data. This dataset is suitable for research directions such as analysis of maritime target characteristics, intelligent recognition, and target tracking. After the data is uploaded, it will be accompanied by a multi-source observation data analysis software (V1.0) for sea targets, which supports efficient data analysis and analysis, providing important support for related academic research and engineering applications. Reference format for this dataset: [1] LIU Ningbo, LI Jia, WANG Guoqing, et al. Sea-detecting Radar Experiment and Target Feature Data Acquisition for Multi-source Observation Data Set of Marine Targets [J]. Journal of Radars, in press. DOI: 10.12000/JR25001.[2] GUAN Jian, LIU Ningbo, WANG Guoqing, et al. Sea-detecting radar experiment and target feature data acquisition for dual polarization multistate scattering dataset of marine targets[J]. Journal of Radars, 2023, 12(2): 456–469. doi: 10.12000/JR23029[3]LIU Ningbo, DING Hao, HUANG Yong, et al. Annual Progress of Sea-detecting X-band Radar and Data Acquisition Program[J]. Journal of Radars, 2021, 10(1), doi: 10.12000/JR21011[4] LIU Ningbo, DONG Yunlong, WANG Guoqing, et al. Sea-detecting X-band radar and data acquisition program[J]. Journal of Radars, 2019, 8(5): 656–667. doi:10.12000/JR19089

雷达海洋探测数据集（Radar sea detection data）——海洋目标多源观测测试数据集，覆盖7类典型海洋目标，包括干货船、液货船、集装箱船、客船、渔船、工程船及搜救船，每类包含4组数据，总计28组数据（ZIP压缩文件），总数据量约42GB。数据内容涵盖多源观测信息，包括雷达HH/VV极化中频/视频回波、可见光与红外图像、目标轨迹（雷达轨迹与船舶自动识别系统（AIS））、气象水文参数（风速、风向、浪高、浪向），覆盖不同海况、天气条件与目标运动状态。该数据集时空分辨率高、数据完整性强，可为海洋目标探测、识别与运动分析研究提供高质量数据支撑。

数据采集设备部署于烟台第一海滩。雷达采用圆周扫描模式开展长期连续观测，并基于实时目标位置信息动态截取局域数据，以优化存储与解析效率，降低冗余数据负载。光电设备工作于被动模式，接收雷达提供的实时目标距离、方位与俯仰信息，并自动调整自身参数，确保多源数据精准对齐。

本数据集适用于海洋目标特性分析、智能识别、目标跟踪等研究方向。数据集上传后将配套发布海洋目标多源观测数据分析软件（V1.0），支持高效数据处理与分析，可为相关学术研究与工程应用提供重要支撑。

本数据集参考文献格式如下：
[1] 刘宁波, 李佳, 王国庆, 等. 海洋目标多源观测数据集的探海雷达实验与目标特征数据采集[J]. 雷达学报, 已录用. DOI: 10.12000/JR25001.
[2] 关健, 刘宁波, 王国庆, 等. 海洋目标双极化多态散射数据集的探海雷达实验与目标特征数据采集[J]. 雷达学报, 2023, 12(2): 456–469. DOI: 10.12000/JR23029
[3] 刘宁波, 丁浩, 黄勇, 等. 探海X波段雷达与数据采集方案年度进展[J]. 雷达学报, 2021, 10(1). DOI: 10.12000/JR21011
[4] 刘宁波, 董云龙, 王国庆, 等. 探海X波段雷达与数据采集方案[J]. 雷达学报, 2019, 8(5): 656–667. DOI: 10.12000/JR19089

搜集汇总

背景与挑战

背景概述

该数据集是一个海洋目标多源观测数据集，包含7种典型海洋目标类型（如干货船、液体货船等），共28组数据，总数据量约42GB。数据内容涵盖雷达回波、可见光/红外图像、目标轨迹和气象水文参数，具有高时空分辨率和完整性，适用于海洋目标检测、识别和运动分析研究，并附带数据分析软件支持。