Urban Waterlogging Monitoring Method Based On Crowdsourcing Model

The flood disaster caused by global warming and urbanization has seriously affected the normal operation of the city and the daily life of the residents.However,the existing urban waterlogging monitoring methods have strict environmental requirements and limited monitoring scope.In order to solve this problem,this thesis proposes a method of urban waterlogging monitoring based on crowdsourcing model,which provides important help for the monitoring of urban waterlogging.The main work includes:(1)Aiming at the problem that task quality,task budget and task real-time are difficult to guarantee at the same time in the process of crowdsourcing task allocation,a multi constraint crowdsourcing task allocation method is proposed.This method first establishes a task allocation model to calculate the credibility of all participants to complete the task;then,based on the level of task participants and task budget,it obtains different allocation combination methods;finally,it assigns tasks to the group with the highest set credibility among all combination methods to execute.Compared with the traditional crowdsourcing task allocation method,it is found that this method not only improves the real-time performance of the task,reduces the task budget,but also improves the quality of the task to a certain extent,which can effectively allocate crowdsourcing tasks.(2)In view of the time-consuming and low accuracy of crowdsourcing data fusion,a method of crowdsourcing data fusion in urban waterlogging is proposed.Firstly,the method combines geohash and DBSCAN algorithm to carry out spatial clustering for the collected waterlogging data.Then,the spatial data and water depth data after clustering are fused respectively.Among them,for water depth data,this thesis proposes a data fusion method based on Grubbs detection,which is used for detection,filtering and fusion.Finally,experiments show that this method can effectively reduce the time of data fusion and improve the accuracy of data fusion results.(3)Based on the above research,the urban waterlogging monitoring system based on crowdsourcing mode is designed and implemented.The system includes mobile terminal and Web terminal,and has the functions of task assignment,employee information management,waterlogging state visualization,waterlogging data collection,participant information registration and cancellation.