Achievement Of Prevention And Control Command System Under The Guidance Of Deep-Learning

As the continuous development and integration of the country’s social economy,the political and economic situation at home and abroad is complex and grim.Therefore,various problems and challenges are confronted by the society.It’s necessary to use information technology to enhance the three-dimensional prevention and control of public security,actively respond to the development of social security,to deeply promote the construction of three-dimensional social security prevention and control system,and to comprehensively improve the social security governance capabilities under the conditions of dynamic informationalization.Based on the design and implementation of the three-dimensional prevention and control command system of National Innovation and Entrepreneurship Base in the city of XXX,this paper analyzes the demands and display methods of the project,and then,divides the system into three major platforms including video surveillance platform,police research and judgment platform,and comprehensive command platform.The system is designed and implemented according to three major platforms.The video platform mainly performs playback control on all video images in the base,labeling video images,playing picture-in-picture,etc.,and augmenting reality on the video images.In the video platform,the TensorFlow Object Detection API object recognition technology in deep learning is applied to mark and recognize objects such as pedestrians and vehicles in the video.Based on this data,the density detection and stay analysis of people’s behavior analysis are realized.The police platform mainly processes and responds to the real-time alarm data.After analyzing the feedback of cases’ coordinates,the police hot icon can be realized,which provides a visual display for the command decision and the police deployment arrangement,and also demonstrates the police’s analysis and research of a certain type of case.In the comprehensive platform,the intelligent transportation’s functions including including vehicle bayonet inspection and control,bayonet vehicle inquiry,traffic light control,face bayonet identification alarm,vehicle trajectory analysis,and induced screen information release of are achieved.Through the cooperation of these main functional modules,the information services are available for the base’s prevention and control command.The mainstream mode architectures of B/S and of C/S are adopted in the technical implementation of the system.The former one is adopted in the police research and analysis platform and the comprehensive command and dispatch platform,and the later one is used in the video surveillance platform.The front-end display mainly uses C# for client development,and VUEJS for Browser expolaration while the back-end business logic is written in Java language,and the SpringMVC framework is used for back-end functional interface development;in video surveillance,based on deep learning,the target recognition is explored by the use of python 3.5.In terms of database,both the combination of Redis,Mysql and Oracle and of the primary database-backup database provide a reliable guarantee for the data security of the system.The three-dimensional prevention and control command system has been put into operation,and has achieved good feedback,which has improved the informationization of the base prevention and control command to a certain extent and increased the work efficiency.