The Role of Digital Generalization in Image Segmentation

The use of remote sensing imagery to obtain land cover and land use maps is a common task in GIS applications. Segmentation techniques identify closed regions in images, producing vector datasets composed by polygons. Since segmentation is a bottom-up technique, the resulting vector datasets are often too detailed. Thus, we need to use generalization techniques to reduce data storage and generate maps with different degrees of detail at different scales. This paper proposes new method for polygon generalization, useful for vector data sets drawn from remote sensing data.