Research On Matching Methods For The Vector Spatial Data With The Same Scale Based On Urban Skeleton-line Network

With the quickening of the urbanization process in our country, the road network and habitation change quickly and largely, which brings large difficulty to keep the map verisimilitude. The main method of keeping map verisimilitude is data updating, while the key technique in data updating is data matching. Taking the two most important features in the urban map—roads and habitations, as the research objects, this paper focuses on the road network matching and habitation matching(both of the two types of matching data are at the same scale) by using the urban skeleton-line network, and the main contribution and creative points are listed as follows:1. Summarizing and analyzing the background and present research about the road network matching and habitation matching. Through discussing the problems existing in the matching process, some new matching methods aimed at the same-scale matching data are proposed based on the urban skeleton-line network, and the connections between roads and habitations are also established at the same time, which can provide a new judgment criterion adding to the matching process of both of the two matching data, thus to help to enhance the matching correctness. This is the essential meaning of building urban skeleton-line network and its advantages in the matching process.2. Research on habitation matching based on the skeleton-line mesh. For the matching of habitation data with location deviations, the habitation matching can be transformed into skeleton-line mesh matching by the one-one reflection relationship between skeleton-line mesh and habitation. Firstly build the dual graph of the skeleton-line mesh by using their connection topology, then the matching results of skeleton-line mesh can be obtained by measuring the centrality similarity of the nodal points in the dual graph. Finally transform the mesh matching results into habitation matching results with the reflection relationship mentioned above. This method solves the matching problems of the habitation data with location deviations.3. Research on the road network matching considering the similarity of the neighboring habitation cluster. This method draws on the experience of principle and characteristics of the human cognitive theory for the unfamiliar surroundings, taking the distribution similarity of the habitations around the road feature as the measurement index. The habitation clusters are determined on the basis of the urban skeleton-line network, and the road network matching results can be obtained by measuring the similarity of spatial relationship and geometric characteristics of the habitation clusters. This method solves the road network matching problems caused by location and rotation deviations.4. Research on avoiding global ergodicity in the road network matching process. A hierarchical and iterative matching method based on road topology classification is put forward.This method classifies the road network into different types. Then the road network is divided into matching hierarchy and unmatching hierarchy based on the number of the different road types, which can guarantee that there are only a few roads in each matching hierarchy. When the road matching begins, it only needs to search the matching objects in the corresponding matching hierarchy including the same road types. This process is moving in circle until the end of the matching by using the iterative operation, which guarantees that the road network is classified dynamically. This method avoids the global ergodicity, raises the search speed and improves the matching efficiency and correctness.5. Spatial Data Matching Test System based on the urban skeleton-line is designed and fulfilled. Taking the multi-source road network and habitation data at the same large scale as the experimental objects, the scientificalness and validity of the proposed theories and algorithms are tested and verified.