Research And Realization Of Intelligent Video Analysis Technology In Unmanned Substation

With the continuous development of information technology and the improvement of the security standard,Intelligent Video Analysis Technology has attracted more and more attention.As an important part of the construction of power systems,smart grid's safe and efficient operation links to the stability and security of the entire power system.And the security monitoring technology plays an important role of the power monitoring technology from the traditional "Five Remote" system till now.This paper presents several effective approaches for the intelligent video analysis technology in unmanned substation.Aiming at the identification of the person in unmanned substation video surveillance,a real-time video-based face recognition method is proposed based on Convolutional Neural Networks(CNN)and CUDA.A 6-layer CNN was built,and the faces in video frames detected by HaarAdaboost will be entered into the CNN.The whole process was accelerated by CUDA.In addition to be more suitable for the actual situation,open-set face recognition was introduced by multistage decision which process the results of the Softmax classifier.Combined the double background model of different learning rates and the finite state machine to realize the detection of the abandoned objects in the surveillance video.And the relative experiments were carried out to verify the performance of the methods.Finally,the real-time video-based face recognition method and the abandoned object detection method were migrated to the GPU-based embedded device Jetson TK1,to complete the intelligent video analysis system of the front embedded deployment mode.According to the experiment,this way gets a good performance,while it can save the transmission bandwidth by only transmit the results of the intelligent video analysis methods which we conducted.