Contour layer extraction from colour topographic map by feature selection approach

Abstract

Contour layer of topographic maps can provide useful information for many vector based GIS applications. A topographic map consists of various semantic layers of overlapping information, differentiated primarily based on colour. Hence, to extract the contour layer from a topomap, colour segmentation algorithms are employed. However due to scanning and inherent features of topomap, colour space selection which provides efficient segmentation results becomes essential. The method proposed uses PCA as feature selection approach to select the colour features of topomap. The clusters are further processed to map to the layers of topomap using spatial properties. © 2011 IEEE.