Design And Implementation Of Intelligent Community Policing System

Urbanization of our country is in rapid development period,complex and varied source of population migration,staff activity area is big,Population is separated from households,grasp the situation is difficult,difficult to obtain useful information,abnormal events are difficult to take the initiative to find,community policing safe hidden trouble,these issues to the public security police alert mining pressure also increased gradually,the traditional mode of police is difficult to meet the demand of modern management,Need to use information technology to improve work efficiency to meet the needs of daily police work.Along with our country aging faster,concerned about the old man lives in the community is also very important,the data shows,the old man more than 80% of the accident life accident reason is caused by fall,if you can find in time the old man fell down,its harm to health can be reduced by more than 85%,in this thesis,based on the existing pedestrian falls algorithm was improved,It is a key function of community governance system to improve the identification efficiency of elderly falls and minimize the injury of elderly people from accidental falls.The progress of modern science and technology has brought strong hardware and software support to the realization of intelligent community policing system.Effectively improve the efficiency of community governance services of police.The specific functional modules of the system are designed according to the actual work needs of police officers’ community governance,and the overall technical framework of the intelligent community policing system is determined.A complete intelligent community policing system is designed and realized,which contributes to the modernization of intelligent policing.The main research work of this thesis is as follows:（1）Taking Q District of W City as the experimental area,appropriate software development framework was selected to participate in the design of functional sub-modules of the police system combined with the actual needs of police officers’ community governance,including improving the fall recognition algorithm,thematic application peer vehicle analysis,abnormal alarm and other key functional modules.（2）the improvement project of the old man fell down recognition algorithm,first by interframe difference method to calculate the test target,using the Open Pose by the SVM classification algorithm is the key to the human body detection algorithm to identify the body posture,can quickly identify the community elderly fall posture,in time to rescue,will fall to minimize the harm of the old man.The test results of MPII Human Pose experimental data set show that the accuracy of the proposed algorithm is 5% higher than that of the Open Pose algorithm.（3）Design and complete the anomaly analysis model of big data,which can realize the anomaly analysis of people and vehicles during the day and at night and zombie vehicles.