Remote Sensing Image Spatial And Temporal Database Research

The remote sensing may obtain a wide range of information with more means, more speed, shorter cycle and less conditions, etc. And with the rapid development of aviation and space technology, the cost of remote sensing data acquisition is lower and lower .So it has been widely applied to land using, environmental monitoring and weather forecasting. However the amount of remote sensing data is very large, moreover, the spatio-temporal data also contain more than one data and there are a lot of historical data. So the storage of remote sensing data is more difficult.The data take more disk space and are not easy to manage. Therefore, how to establish an effective spatio-temporal raster data model to express the spatial and temporal characteristics of geographical entities and how to compress the amount of remote sensing data to save disk space have been attracted much attentions in the field of current raster spatio-temporal database .This article first briefly introduced the background and significance of the research topic, analyzed the foundation of spatio-temporal raster data modeling and several common space of raster data model, and mentioned the design idea and the concept model of double ground-state amendatory model. Secondly, spatio-temporal raster database has been designed from the conceptual and logical. Then we ameliorated the fast dynamic linear quad tree coding algorithm on the basis of analysis of the current raster data compression method commonly used. The improved algorithm is applied to the two-dimensional run-length dynamic encoding, which is adopted in double-ground-state with amendments. Then we have analyzed and researched on how to identify the differences between the phases of raster data. The analysis and cumulative difference algorithms are improved. Finally, we have designed and built a grid space-time database on the relational database SQL Server 2000. The spatio-temporal data management system has been developed and implemented based on double ground-state amendatory model,using Visual Studio.Net 2005 and C# language.Besides, taking the three remote sensing image data of Liaocheng 1987-2005 as an example, we made analysis about the idea of double ground-state amendatory model and the proposed algorithm.The Validation result shows that, compared to the general spatio-temporal raster data model and compression algorithms, double ground-state amendatory model introduced in this paper is more reasonable on organization and management of raster data. Two-dimensional dynamic compression stroke of the raster data compression algorithm is faster and more efficient for the compression of spatio-temporal raster data.On this basis, the system has been built, which improve the speed of raster data's input, retrieval, query and update.