Evaluation of Cartosat-2E Data for Large-Scale Urban Mapping

Abstract

Large-scale maps are a prerequisite for sustainable management of urban areas for which very high-resolution satellite (VHRS) imagery can be effectively utilized. PAN sharpened (merged) orthorectified product of Cartosat–2E satellite launched by ISRO (0.6 m spatial resolution) has been evaluated for 1:4000 scale mapping in Dehradun and Rishikesh, India. Root mean square error (RMSE) of Cartosat-2E data is found to be 29.03 m in Dehradun and 28.15 m in Rishikesh. However, after the induction of GCPs collected from the Differential GPS (DGPS) survey, RMSE error reduced to less than 1 m, which is acceptable for urban mapping as per atal mission for rejuvenation and urban transformation (AMRUT) design and standards. It is observed that system-level improvements in data acquisition and the use of high accuracy GCPs can assist in harnessing the potential of Cartosat-2E for large-scale urban mapping at 1:4000 scale. It is also found that Cartosat 2E data can be utilized for identification of a large number of subclasses (~ 39) of urban land use land cover and building footprints as per the AMRUT design standards which have a great potential to reduce project cost and volume of human resources for large-scale urban mapping as compared to conventional ground-based methods.