Multi Feature Extraction And Hierarchical Intelligent Recognition Model For Urban Road Data Based On Map Generalization

Automatic map generalization has been a hotspot of scholars, but also one of the challenges in cartography and most creative research in the field, where the road network is integrated comprehensive map of the important content, overall quality of the road network decided to map overall quality. After analyzing and studying the interrelated literature at home and abroad, the author researched and explored the physical topology reconfiguration、topology preprocessing and comprehensive road network. This paper studies the following aspects:1. Many topology based algorithms are used to deal with points, lines, polygons and other entities in the map generalization, and the topological entity reconstruction algorithm is proposed to reconstruct the points, lines and polygons.2. In view of the problem of the error of spatial data in the spatial data found in the automatic generalization, the author proposes a fast algorithm to identify and deal with such errors spatial relationships. In order to study the spatial relationships of spatial data errors, use topological relationships to express their spatial relationships, according to the characteristics of spatial data distribution patterns to identify different types of topology errors, and take the appropriate algorithm for different modes for processing.3. Based on the topological entities reconstruction and topology preprocessing, the automatic extraction of geometric characteristics of the road stroke, topological features, spatial distribution and semantic features, and further proposed grade intelligent recognition model to calculate the road grade stroke in the road network, Finally, according to the road level to complete the stroke automatically integrated road network in the path selection. Simplification of road and road shift are proposed new algorithms to meet different types of data and the different needs.By the series of independent research and development platform for software programming algorithm module corresponding experiments, applications, verify the proposed method on the basis of typical algorithms and study the correctness and reliability.