| With the rapid development of science and technology, a number of different geographic data been collected, the geographic information system with the effect of storage, processing and analysis is outstanding increasingly. Current processing of geographic information system is mainly based on static 2-D digital maps. On the complex multi-dimensional spatio-temporal data, a more visual, interactive environment is needed to process, analysis and display data.The main research content of this paper includes:(1)Taking SOM neural network as method for visual analysis of spatial-temporal data.SOM neural network is a kind of neural network for high dimensional data clustering, which not only can maintain the topological structure unchanged, but also convert high dimensional data to low dimensional display by way of abstracting generalization;(2)By using the method of multiview, the spatio-temporal data visualization and analysis model is set up, and the visual expression and visual analysis of spatio-temporal data from multiple angles are analyzed;(3)The visualization of spatial-temporal data helps us to analyze the temporal and spatial changes of spatio-temporal data.Taking NASS site’s United States agriculture survey data as data source,the paper established agricultural spatio-temporal database. With Visio studio 2010 as the development platform and c # language programming,according to agricultural production spatio-temporal data, the visualization interface of agricultural spatio-temporal data is designed. The interface adopts collaborative multi view expression to show evolutionary process of spatio-temporal data under different combination of classification, which users can choose different combination of attributes and time sequences. The system adopts SOM method for visual analysis of spatio-temporal data, shows various types of data labels by the way of cellular figure, and provides its space-time distribution and evolution by the method of space-time cube, ultimately. |
PDF Full Text Request