Research On Matching And Connection Organization Method Of Multi-scale Road Data

Geographical entities are unique in the real world.However,due to the difference of expression scale,the expressions of the same geographic entity in different scale spatial data are different.At the same time,the spatial data of the same geographic entity in different scale spatial data are lack of vertical correlation across the scales.It disconnects the correlation of the same geographical entity in different scales and is not conducive to the comprehensive application of multi-scale spatial data.Road is an important infrastructure,road data is an important spatial data.It is closely related to peopled daily life,and there are various kinds of road spatial data.China's basic geographic information spatial database contains abundant road data.There is a lack of correlation between different scale road data expressed in layers,which cannot support cross-scale analysis and application.It is of great significance to improve the application value of multi-scale road spatial data by associating and organizing the multi-scale road spatial data which expressed in layers and establishing the correlation relationship between the same road entities in different scales.Therefore,this paper takes the multi-scale road network data as an example to discuss the matching and connection organization method of the same road entities.The main research contents are as follows:(1)The research status and the existing problems of multi-scale data integration organization,multi-scale correlation relation model and road data matching method are analyzed.The differences of the same road entities in different scales are analyzed,and the causes and the types of differences of the same road entities are summarized.(2)A method of multi-scale vector road data matching which considered the neighborhood is proposed.First of all,based on the different types of the same road entities in different scales,the characteristic factors of road matching are summarized,and the evaluation index and calculation method of road entity similarity are designed according to the characteristic factors.In order to effectively identify the correspondence between the same road entities in different scales,a multi-scale vector road matching method which considered the neighborhood is proposed.When matching,the road Voronoi diagram is firstly constructed as the road matching unit and used to select the candidate matching data set.Then,geometric similarity is mainly considered,topological similarity and semantic similarity are secondary considered,and the multi-scale vector road data are preliminarily matched according to the constraint rules.Finally,the unmatched roads are combined with the matched road in the neighborhood before matching,so as to realize multi-scale vector road data matching.(3)Multi-scale road data connection organization model is proposed.Based on the data organization method of Geodatabase model,the multi-scale road data connection model is designed.In order to realize the corresponding representation of the same road entities in different scales,the processing method of one-to-one correspondence of road entities with the same name is proposed.According to different matching results,it formulates the processing rules of the one-to-one correspondence relation,and then uses the multi-scale road data connection model to organize the processed road data.It makes the same road entities in the adjacent scale road data correspond one to one.Finally establishes the one-to-one correspondence relation of the same the road entities in different scales.(4)Based on the theoretical research results,the system of multi-scale road connection organization is designed and developed.The system has the following functions:road data preprocessing,structuring road matching unit,multi-scale road entity matching,multi-scale road data connection organization,searching the multi-scale road and multi-scale path.The 1:10,000,1:50,000 and 1:250,000 road network data of Nanjing are used as experimental data.Firstly,the same road entities in the adjacent scale are matched,and then a one-to-one correspondence connection is established for the same road entities after matching,so as to realize the connection organization of multi-scale road data.Experimental results show that the multi-scale vector road matching method which considered the neighborhood has higher matching accuracy and better robustness.And the multi-scale data connection organization model can effectively realize the multi-scale road data connection organization.