Vectorization and spatial query architecture on island satellite images

Vector data structure produces smaller file size than raster image because a raster image needs space for all pixels while only point coordinates are stored in vector representation. Raster-to-vector conversion, i.e. vectorization, has key functionalities in Geographic Information Systems (GIS). This study presents a technique converting raster satellite island images into vector data as polygons by using edge-detection algorithms. This enables storing satellite images as vector objects into spatial database systems and utilizing the advanced built-in spatial functions and queries developed by the spatial database community. The effectiveness of the technique is demonstrated by applying on a system detecting spatiotemporal changes in a real-world island satellite image.