The Design Of Intelligent Video Monitoring System For Substation

With the comprehensive transformation of power sector, in order to improve the safety production efficiency, the field to be monitored in transformer substation has become more and more widely, in addition to the security monitoring field also includes of substation equipment instrument and environment. So the number of video channels becomes more and more large, if all the video flow to monitoring center for monitoring staff to analysis, on the one hand the labor cost is too large, besides the useful information accounted for less than 1%, which will result in the waste of network resources. On the other hand, the low efficiency of manual control,the visual fatigue and human slack easily lead to accidents which are not only unable to capture but also cannot be prevented.To surmount these disadvantages above, this paper designs a set of substation intelligent video surveillance system. The system can extract the effective information from video stream automatically almost without human intervention.And it can alarm for the behavior or phenomenon within the set rules. The main content of this paper includes the following parts:Firstly, introducing the related technologies for the system design and implementation, including image preprocessing, moving object detection technology and other related computer technology.Secondly, the paper discusses the application of video intelligent analysis algorithm in substation safety production and designs algorithms for some applications, including people monitoring, articles monitoring and instrument monitoring. Focus research:(1)Trip detection algorithm. Based on the relationship between three points and line, a new algorithm of one-way trip detection was put forward. The algorithm can avoid the wrong alarm when the person wander beside the line, which improve the rate of correct recognition.(2)Illumination invariant feature extraction algorithm for face recognition. Concerning the illumination changes on the influence of the recognition rate in face recognition, a new face recognition algorithm for varying illumination environment was proposed according to the classicillumination invariant feature representation algorithm, which is based on Local Standard Deviation and its weighted form.(3)Automatic reading algorithm for pointer type meter. Modeling Concerning the illumination changes and the low speed of pointer instrument detection, which show it difficult to achieve readings fast and accurately according to existing algorithms, a new algorithm is proposed to solve this problem based on scan line processing.Finally, based on software development technology, an auxiliary intelligent video surveillance system for substation safety production is developed, which can realize the function of the algorithm design.