Compression And Multi-scale Expression Of Spatial Data Based On GIS

The spatial data arc the blood of the geographical information system (GIS) . In fact, the whole GIS surround gathering, process, store, analyse and representation of spatial data. With constant enlargement of GIS application domain. People need abundant and various spatial data more and more. However, spatial data of vast range and high accuracy need the memory space is very enormous. Furthermore, because of bandwidth restrictions of network, multi-user networked serve is too difficult to finished. Especially, for some GIS, such as flood prevention, traffic, and so forth, need transmit information fleetly. Hence, raising the transmission speeds of datum is necessary. In addition, mass spatial data render the efficiency of the data inquiry and management is limited. Therefore, the compression technology of spatial data has become necessary. It is one of the key problem of practicability and networked serve of GIS. The day-by-day improvement of demand level of GIS has brought to a large amount of demands of analysis and expression of the spatial data in multi-resolution. People have demand observe, understand and describe the geographical phenomenon in different resolution and spatial scale. Namely, analyseing, dealing with and expressing the spatial data in multi-resolution.The main content of this paper is to design compression and expression in multi-resolution of the vector data and Digital Elevation Model (DEM) . It also includes the compression of the image database. In this paper, wavelet, scalar quantity, vector quantity and artificial neural networks have been fully utilized. Besides, the application of these methods for spatial data compression and expression in multi-resolution has been researched and disussed deeply.At first, the concept and development of GIS, the content and datum structure of the spatial data and the basic theory of wavelet analyses are introduced briefly. Emphatically, analyse the space and frequency distribution characteristics of wavelet coefficient of image.Secondly, compression and expression models in multi-resolution of vector data are recommended. Concrete compression model and algorithm have been offered. Theproblems that distortion problem of border of compressed the vector data and shift of characteristic points have been solved.Then, compression and expression in multi-resolution of DEM has been researched.Application of wavelet transform is introduced. Furthermore, wavelet coefficient of DEMis analyzed and summarized detailedly. The optimal scalar quantity in wavelet domain isdesigned. The model of multi-resolution based on wavelet multi-resolution analysis(MRA) is constructed.Finally, contraposing to the magnanimity image database in the spatial database, the compression method of wavelet combined with vector quantization is put forward. Moreover, the basic principle and general algorithms of vector quantization and analyzes the technology characteristics of vector quantization in wavelet domain are introduced. Especially, the key technology of vector quantization is discussed and two kinds of new design method of codebook are put forward.To the problems of Compression and multi-scale expression of spatial data, wavelet MRA and scalar quantity, vector quantity based on wavelet offer the new train of thought and method. The space memory and transmission problem of spatial data are settled better. Besides, requisition for the spatial data of multi-resolution is meet. Therefore, they have stronger practicability and extensive application prospect.