京张高铁智能视频分析数据集

视频分析数据集，主要由站内异常行为检测，包括站台越线、端部入侵、扶梯逆行、排队长度和人群密度等，由智能算法自动检测异常事件，形成相应的报警片段进行存储。
智能视频分析报警片段数据，针对各个客运站的不同异常行为进行检测，以“车站-报警类型-时间”的方式存储，每个视频片段包括异常事件地点、类型以及报警图片和事件发生的前5秒和后5秒的视频片段。数据量57MB。

This video analytics dataset is primarily designed for in-station abnormal behavior detection, covering scenarios including platform boundary crossing, end intrusion, escalator reversal, queue length assessment and crowd density detection. Abnormal events are automatically detected via intelligent algorithms, and the corresponding alarm clips are generated and stored.

This intelligent video analytics alarm clip dataset is tailored for detecting various abnormal behaviors in different passenger stations, and is stored in the format of "station - alarm type - time". Each video clip contains the location and type of the abnormal event, along with the alarm image and video clips of 5 seconds prior to and 5 seconds after the occurrence of the event. The total size of the dataset is 57 MB.

搜集汇总

背景与挑战

背景概述

该数据集是京张高铁智能视频分析数据集，专注于站内异常行为检测，包括站台越线、端部入侵、扶梯逆行、排队长度和人群密度等场景，通过智能算法自动识别并存储报警视频片段。数据集包含11个文件，总大小57.02MB，格式为mp4和docx，来源于国家重点研发计划项目，旨在支持高铁智能化服务技术示范。