RoadSens-4M Dataset

The <b>RoadSens-4M Dataset</b> provides a multimodal collection of sensor, video, weather, and GIS data designed to support research in intelligent transportation systems, road condition monitoring, and machine-learning-based anomaly detection. This dataset integrates synchronized smartphone sensor data (accelerometer, gyroscope, magnetometer, GPS) with video annotations, weather, and geospatial information to accurately identify and classify road surface anomalies, including bumps, potholes, and normal road segments.<b>The dataset comprises 103 data sessions organized in a hierarchical structure to facilitate flexible access and multi-level analysis.</b> It is divided into four main components: <b>Raw Data</b>, <b>Combined CSV with GIS and Weather Data</b>, <b>Isolated Data</b>, and <b>GIS Data</b>. Each session folder contains all corresponding sensor CSV files, including both calibrated and uncalibrated readings from the accelerometer, gyroscope, magnetometer, barometer, compass, gravity, and GPS sensors, along with annotation and metadata files. Within every session, a dedicated <b>camera subfolder</b> holds annotation data and a text file linking to the corresponding video stored on Google Drive, allowing researchers to access complete recordings without manual segmentation.The <b>merged CSV files</b> combine synchronized sensor, GIS, and weather information (temperature, humidity, wind speed, and atmospheric pressure) with a sampling interval of <b>0.01 seconds</b>, ensuring high temporal resolution. The <b>Isolated Data</b> folder further separates normal and anomaly samples to enable focused comparative analysis, while the <b>GIS Data</b> folder contains QGIS and elevation files for spatial and topographical visualization.This well-structured organization ensures seamless integration of sensor, video, geographic, and environmental data, supporting efficient navigation and in-depth multimodal research. The <b>raw data</b> are hosted separately on Google Drive and can be accessed via the following link:<br>🔗 https://drive.google.com/drive/folders/16tRSgXy6bjgIcJZzdw3U5unw7jpsKAHB?usp=drive_link

<b>RoadSens-4M 数据集</b>是一套多模态数据集合，涵盖传感器、视频、气象及地理信息系统（Geographic Information System, GIS）数据，旨在为智能交通系统、道路状态监测以及基于机器学习的异常检测相关研究提供支撑。该数据集将同步后的智能手机传感器数据（加速度计、陀螺仪、磁力计、全球定位系统（Global Positioning System, GPS））与视频标注、气象及地理空间信息进行整合，可精准识别并分类路面异常状况，包括颠簸、路面坑洼与正常路段。<b>该数据集共包含103个数据会话，采用层级化结构进行组织，以支持灵活访问与多维度分析。</b> 它被划分为四大核心模块：<b>原始数据（Raw Data）</b>、<b>融合地理信息系统与气象数据的组合CSV文件（Combined CSV with GIS and Weather Data）</b>、<b>分离式数据（Isolated Data）</b>与<b>地理信息系统数据（GIS Data）</b>。每个会话文件夹内均包含所有对应的传感器CSV文件，涵盖加速度计、陀螺仪、磁力计、气压计、罗盘、重力传感器以及全球定位系统传感器的校准与未校准读数，同时附带标注文件与元数据文件。在每个会话内部，设有专属的<b>相机子文件夹（camera subfolder）</b>，其中存储有标注数据，以及一个指向存储于谷歌云端硬盘（Google Drive）对应视频的文本文件，方便研究人员无需手动分割即可获取完整的录制数据。<b>合并CSV文件（merged CSV files）</b>将采样间隔为<b>0.01秒（0.01 seconds）</b>的同步传感器、地理信息系统与气象数据（温度、湿度、风速与大气压强）进行整合，确保了极高的时间分辨率。<b>分离式数据（Isolated Data）</b>模块进一步将正常路段样本与异常样本分离开来，可支持针对性的对比分析；而<b>地理信息系统数据（GIS Data）</b>模块则包含用于空间与地形可视化的QGIS软件文件与高程数据文件。这种结构清晰的组织方式实现了传感器、视频、地理与环境数据的无缝整合，可为高效导航与深入的多模态研究提供支持。<b>原始数据（Raw Data）</b>单独托管于谷歌云端硬盘（Google Drive），可通过以下链接访问：<br>🔗 https://drive.google.com/drive/folders/16tRSgXy6bjgIcJZzdw3U5unw7jpsKAHB?usp=drive_link