An Intelligent Video Analysis System of Power Scene Anomaly Based on Convolutional Neural Network

As a branch of computer image vision technology, video intelligent analysis has been widely used in the field of security, which plays an important role in operation monitoring, fault analysis, early warning automation and information processing in the power grid industry. In order to improve the low efficiency of the artificial patrol mode of the existing substation video surveillance system, this paper designs an intelligent video analysis system for power scene anomalies based on convolutional neural network. Firstly,the system architecture is analyzed. Secondly, the original surveillance video material is extracted. Then, the video cloud storage platform is used for storage. At the same time, the hidden danger behaviors are classified and analyzed and marked. Finally, the proposed method is applied to feature extraction, model training and application analysis to realize video intelligent early warning analysis. Based on convolutional neural network algorithm, the video analysis ability is further improved, which greatly improves the quality and efficiency of substation video monitoring, and has a certain promotion prospect.