Research Of Dynamic Road Generalization Model And Methods Based On Trip Mode

Ample accomplishments of the map generalization have been achieved in its theory,model and method after several decades'development.However,for now,digital map generalization is mainly based on generalization in static environment and offline generalization.Maps that produced by traditional generalization techniques are applied to a wide range of different tasks,which can not meet users' individualized needs for geographic information.With the development of network technology and artificial intelligence technology,internet and mobile map that are highly interactive are becoming more and more popular.At the same time,research gradually turned their focus to the mining of semantic information as well as users' individual information which then lead to the new requirement for map generalization so as to provide any time,any place and any scale geographic information according to users' tasks and context information.Evidently,offline generalization and multi-scaled database are not available under this circumstance.As one of the important context information,trip mode relates tightly to road generalization.Users have different requirements of geospatial information,thus,maps with appropriate scale and appropriate amount of information must be provided in order to meet users' road cognitive demands under diverse trip modes.However,the target scale of traditional generalization is specified,and generalization area is certain while trip mode varies a lot in mobile environment,thus,a dynamic road generalization model should be built to solve these problems.To design such a model,several aspects such as selection of generalization operators and algorithms,set of parameter,match of generalization knowledge and control of generalization process must be taken into consideration.Details are discussed in this paper,specific contents are as follows:1.This paper analyzed user's cognitive characteristics of road and discussed the scales which could meet users' road cognition features based on different trip modes.And we summarized the impact factor of road generalization of road based on trip mode and analyze them.2.The availability of road generalization operators in dynamic environment is analyzed from the perspective of generalization complexity while applicable scale of operators that meet the condition is analyzed too.The availability of road generalization algorithms in dynamic environment is discussed from the perspective of algorithm complexity and quality of generalization.The relationship between algorithm parameters and scales is explored.3.This research described acquisition method and formalization method of road generalization knowledge,and established road generalization-related knowledge rule.A conceptual model of dynamic road generalization based on trip mode is proposed,and controlling mechanism of generalization process used in this model is discussed.4.Based on the model proposed,a prototype of dynamic road generalization for mobile environment has been designed.Two selection algorithms based on attributes and mesh density threshold and one simplification algorithm based on Douglas-Peucker were implemented to execute the dynamic road generalization based on trip mode.Based on the result of case,the feasibility of the dynamic road generalization model is proposed.