Study On Quality Evaluation Method Of Crowd Sourcing Geographic Based On No Reference Data

In the public information of Crowd source,the user collects and produces geographic information data in an open mode as a volunteer,which breaks the way of obtaining traditional geographic information data and provides a rich data source for geographic information data.Crowd source geographic data is generally provided by non-professional volunteers.The quality of data has become an important factor in the wide application of public data.Since the geographic information data that can be used for quality evaluation has high acquisition cost and long update time,The quality evaluation of the public source geographic data is subject to certain restrictions.To this end,this paper proposes an OpenStreetMap(OSM)with the most representative geographic information in the non-reference data scenario,and proposes a quality evaluation method based on the professional level of volunteers and data editing history.The specific research contents of the thesis are as follows:(1)A quality assessment method based on the professional level of volunteers was proposed.The method uses the index weighted model to evaluate the quality of OSM data.Firstly,it extracts four indicators of contribution degree,repeat modification degree,activity degree and concentration degree from the perspective of volunteer professional level,and then selects the median quality of all the contribution data of volunteers as the volunteer.The data quality level of the contribution is finally determined by linear regression to determine the weight of the experimental results of each indicator,and the calculation formula of the data quality evaluation is obtained.(2)A quality assessment method based on the historical version of OSM is proposed.This method considers that the historical version of OSM will have an impact on the final version data quality.Firstly,delete the historical version with normal modification and only attribute modification,and select the data with the number of versions greater than 3,then select the intermediate area method to calculate the difference between each version and the final version respectively.Finally,calculate the final version by weighting the version number.The quality evaluation value of the version data.(3)An OSM data quality evaluation method based on point offset is proposed.The method evaluates the quality from the point of view of the node of the data and the point of mass point.Firstly,the historical version with normal modification and only attribute modification is deleted,and then the number of nodes of the remaining version data is determined.If the number of nodes is the same,each method will be The center of the corresponding node is taken as the real node,and the data composed of the node is taken as the real value.Finally,the similarity between the final version and the generated real value is calculated as the data quality evaluation value;if the number of nodes is inconsistent,the data center of each version is selected.The center is used as a reference point,and the distance between the centroid point of the final version and the reference point is taken as the data quality evaluation value.