Design And Implementation On Multi-scale Adaptive Transformation Model For Spatial Data

Multi-scale representation of spatial data has become one of the most important andforefront issues in geography information science. Scale transformation is one of the mainprocesses of the multi-scale representation of spatial data. The ideal multi-scale transformationshould be entity-oriented, continuous and dynamic. In order to meet different needs of the usersto scan the spatial data, GIS had to store the spatial data in a variety of different scales and levelsof detail，so that would have a lot of data redundancy and the corresponding series of drawbacks.Therefore, you need to integrate into the concept of object and scale in spatial data modeling，regard the real-world object as the object of cognition, and integrate the scale transformation tothe data modeling as a thinking way and a thinking processes to cognize and simulate the world，and then to go to establish an object-oriented, scale-dependent, dynamic data model. Based onreading and analyzing domestic and international literature and information，from the angle oftheories and fulfillment，multi-scale transformation of spatial data is researched. The main workis presented as following：1. Summarizes the space data multi-scale transformation of key technologies of thedomestic and international research progress. The perspectives of multi-scale data model，datastructure and adaptive visualization are introducted，and some problems are found and analyzed.2. Introduces the basic concepts and basic theory with related to scale transformation, andsummes up four patterns about scale transformation which are used at present. These willestablish the theory and technology foundations of spatial data multi-scale transformation.3. A conceptual framework is designed for multi-scale adaptive transformation of spatialdata. The large-scale spatial data is divided into two categories: the matching information and theincremental information. Hence the multi-scale transformation process consists of two steps: themulti-scale transformation of matching information and the multi-scale transformation ofincremental information.4. The multi-scale adaptive transformation for the matching information. This comes intotwo main parts: the multi-scale adaptive transformation of the simple features based on fractaltheory; and the multi-scale adaptive transform of the complex features based on the static anddynamic binding Model.5. The multi-scale adaptive transformation for the incremental information.To realize thecompression and repetition of the incremental information, the multi-scale R-tree and real-timeincremental transformation model are undertaken.