Extending an open source spatial Database with geospatial image support: An image mining perspective

摘要: The nature of vector data is relatively constant, and it revised less frequently as compared to remotely sensed earth observation data. Remote sensing images are being collected nowadays every 15 minutes from satellites such Meteosat. In the coming years, very high spatial resolution expected be available freely frequently. Integrated GIS remote analysis methods have ability incorporate different sources find attribute associations patterns change for knowledge discovery detection. GIS-based DEM overlayed with image results taken up in a further processing analysis. A platform required efficiently store, retrieve manipulate layers just like other hybrid GIS/RS principle, databases most suitable candidates platform. Our work aimed investigate open source database PostgreSQL/PostGIS (PG/PG) platform, provide solution support an overall framework integrated This definitely beyond storage retrieval databases. requirements libraries were extensively studied support. TerraLib library was proposed, analysed extend PG To demonstrate application developed this study, Meteosat Second Generation (MSG) larger part Europe extracted ITC receiver. An programme written construct time series PG/ DBMS. mining detect clouds time-series stored using conceptual schema. For this, extensive study carried out. statistical method based on principal components adopted extract cloud features Netherlands Using research detection case application, various scenarios conducted top DBMS technology. extremely useful studying spatio-temporal phenomena seasonal or long intervals region-based studies where regions by