基于三帧间差分的电厂监控异常检测方法

Anomaly Detection Method for Power Plant Monitoring Based on Three Frame Difference

针对人为检测电厂中视频监控出现异常情况的问题,提出一种基于三帧间差分的电厂监控异常检测方法。首先,将电厂监控视频进行提取获得所有视频帧;其次,采用帧间差分方法并设置阈值进行异常帧提取并将其显示。根据某一段异常监控视频进行数据验证方法的有效性并通过该方法得到该段视频的最佳阈值,所得设置将提高监控异常检测及处理的速度,使用者可以直观检查出异常情况并进行相应的补救措施,显著减少人工管理成本,能够有效提升系统的可用性。

A power plant monitoring anomaly detection method based on three frame difference is proposed to address the issue of human detection of anomalies in video surveillance in power plants. Firstly, extract the monitoring video of the power plant and obtain all video frames;Secondly, the inter frame difference method is used and a threshold is set to extract abnormal frames and display them. The effectiveness of the method is verified by data validation from a certain abnormal monitoring video, and the optimal threshold of the video is obtained through this method. The settings obtained will improve the speed of monitoring anomaly detection and processing. Users can intuitively detect the abnormal situation and take corresponding remedial measures, significantly reduce the cost of manual management, and effectively improve the availability of the system.