Coastal mapping and change detection using high-resolution IKONOS satellite imagery

摘要

Shoreline mapping and shoreline change detection are critical in many coastal zone applications. This paper presents results of the semi-automatic mapping of a coastal area of Lake Erie using 4-meter resolution multispectral (XS) and 1-meter resolution panchromatic (Pan) IKONOS Geo stereo images. An overview of the geometric processing of IKONOS imagery is presented. Then, the latest results of our IKONOS geopositioning accuracy improvement efforts are reported. Subsequently, methods and results of automatic generation of a Digital Elevation Model (DEM) and an orthoimage are outlined. Finally, a novel approach for the automatic extraction of a 3D shoreline from IKONOS images is described. Test results show that the proposed approach is capable of extracting shorelines from IKONOS images with little human interaction. The accuracies of the extracted shorelines from 4m XS and 1 m Pan stereo images are estimated to be 8.5m and 2-3m, respectively.