Research On The Key Quality Control Technology Of Intelligent Transportation System Data

In recent years,traffic management department and travelers are making higher demands to traffic information service.However,real-time and accurate traffic information is based on high quality dynamic data.The current study of Intelligent Transportation System data quality control mainly focus on the innovation of method and algorithm model,a set of standardization process which can guide the users to operate is abcense until now.In order to solve the problem above,the quality problems that exist in the various types of intelligent traffic detection data are studied in this thesis.The main work and results of this thesis are as follows.First of all,based on the in-depth analysis of the domestic and foreign literature,the existing traffic data quality control methods are summarzied from perspectives of errouneous data identification and repair,filling missing data and abnormal data identification and repair.The principle and system composition of vedio,loop,microwave and RFID traffic information collection technology are introduced in detail.Secondly,the data quality control methods of video,loop,microwave and RFID detection data are systematical sorted.The video traffic data processing mainly includes video image pretreatment,video vehicle data acquisition,video vehicle data quality control,video traffic flow parameter acquisition at multi-time period and video traffic flow parameters quality control.Loop detection data processing mainly includes loop pulse signal preprocessing,loop traffic flow parameter acquisition at multi-time period and loop traffic flow parameters quality control.Microwave detection data processing mainly includes maicrowave signal preprocessing,microwave traffic flow parameter acquisition at multi-time period and microwave traffic flow parameters quality control.RFID detection data processing mainly includes RFID original data quality control,RFID traffic flow parameter acquisition at multi-time period and RFID traffic flow parameters quality control.In addition,dynamic threshold estimation model based on Turkey smoothing method is used to identify the the abnormal data.The abnormal data is repaired by the exponential smoothing method,statistical correlation method and linear regression analysis method,and the mean absolute percentage error,the mean absolute error and the root-mean-square error are used to evaluate the accuracy of the three methods the result shows that multiple linear regression analysis has the highest repair precision.Finally,the characteristic of video,loop,microwave,and RFID actual sampling data is analysed in this paper,the problem data is indentified.The appropriate data quality control methods based on the principle of practicality and effectiveness are selected,and the quality control process of video,loop,microwave and RFID traffic detection data is given.