Study On Collaborative Generalization Of Spatial Object Based On Image And Vector

Provided with automatic generalization function of spatial object , GIS not only greatestly manifest the value of spatial foundation data , but also protect investor's benefit and satisfy various,multi-level,the multi-positions GIS's application . So it has the significant economic efficiency and application prospect . Along with the thorough theoretical research and application of automatic generalization of spatial object, study of collaborative generalization of the multi-source data has more and more attracted people's attention . Collaborative generalization is an effective method to resolve the following problems, unclear partial environment condition in generalization process, weak applicability of algorithm, quite a lot exception and the difficulty of use the knowledge in algorithm . The main aim and content of this paper is to use the general computational method of vector data based on image and vector by the man-machine interaction . That is to say, extracting approximate boundary of spatial object from image corresponding to the goal scale, we can get the generalization result of spatial object through calculation based on primitive vector data and approximate boundary . First of all, this paper mention the theory of the traditional settlement's generalization and the thought of physical generalization . And the characteristic of image data in gain process emphatically has been analyzed . In view of the expression request of ground-object element in map, detail degree and figure's evolutionary process of element in image in different resolution have been discussed which indicate the feasibility of image be used in generalization .Secondly, it was introduced that the formulary of calculating model and the method of pretreatment about the boundary purification . Selecting some representative buildings and a part of block, through linking variation process of its approximate boundary in image, using primitive vector data's geometry information and purifying its boundary through calculating model, we can achieve map generalization by the man-machine interaction . Simultaneity, several influence factors that affected the boundary purification briefly have been analyzed . The results of the analysis of the experiment indicate that the method has the certain validity to solute the question of generalization .Finally, the author draws a conclusion and puts forward insufficiencies which in the research as well as further study suggestions .