图像和雷达一体化融合数据

以2020年在甬台温高速公路龙门架上，2022年在西安市二环快速路上，使用相机和毫米波雷达固定好位置采集得到。摄像头采集的帧率为10帧/秒，毫米波雷达采集的帧率为10帧/秒。数据包含黄昏和晴天两种光照类型，经过数据清洗、数据标注、时空同步等数据处理工作。最终得到雷视融合感知的车辆轨迹数据。数据采集于真实的高速公路场景，利用自助研发算法，具有较高的可靠性和质量。数据量1.2GB。The dataset was collected in 2020 on the gantries of the Yongtaiwen Expressway and in 2022 on the elevated Second Ring Road of Xi’an using fixed-position cameras and millimeter-wave radars. Both the cameras and the millimeter-wave radars operated at a frame rate of 10 frames per second. The collected data covers two lighting conditions: dusk and sunny weather. After undergoing data cleaning, annotation, spatiotemporal synchronization and other data processing procedures, the final vehicle trajectory data for radar-vision fusion perception is obtained. The data was gathered from real-world highway scenarios, developed with self-developed algorithms, and features high reliability and quality. The total size of the dataset is 1.2 GB.

搜集汇总

背景与挑战

背景概述

该数据集是图像和雷达一体化融合的车辆轨迹数据，采集自真实高速公路场景（包括甬台温高速公路和西安市二环快速路），涵盖黄昏和晴天两种光照条件。数据通过相机和毫米波雷达同步采集，经过清洗、标注和时空同步处理，专注于雷视融合感知技术，适用于交通运输工程领域的研究和应用。