Dataset of Abnormal Events of Large Machinery Uncontrolled Movement

The current industrial field faces safety hazards caused by uncontrolled large machinery, especially in the detection of abnormal behavior in dynamic scenarios with stackers, cranes, and similar equipment, requiring effective solutions. Current monitoring systems heavily rely on manual inspections, which are inefficient and prone to missed or false reports, failing to provide timely responses. To address this technical challenge, this dataset offers a large volume of video footage of machinery uncontrolled events in real scenarios, aiming to enhance the intelligence of automated monitoring systems. During data collection, high-resolution cameras were used for real-time recording at industrial sites, ensuring the authenticity and diversity of data. We implemented multiple rounds of annotation and expert review mechanisms to ensure data quality, and data is stored in MP4 format to facilitate subsequent analysis and processing.