Using LIDAR-based DEM to orthorectify Ikonos panchromatic images

Abstract

Recent development in remote sensing technologies has stimulated a great interest in its application in large-scale mapping. For example, the Ikonos satellite images, available since 2000, have an improved resolution of 1 m that afford mapping of a lot of ground objects from the satellite images. However, the use of these images in high-accuracy mapping applications has been limited since the sensor model parameters are not yet released. On the other hand, Light Detection and Ranging (LIDAR) is a fast method for sampling the earth's surface with a high density and high accuracy point cloud that is used to generate high density and high accuracy Digital Elevation Models (DEMs) and DSMs. Integrating both technologies makes it possible to provide reliable and automatic solutions for large-scale mapping applications, 3-D visualization, GIS and change detection. In this research a 1-m resolution LIDAR-based DEM is used to orthorectify a single Geo-panchromatic Ikonos image. The LIDAR-based DEM is first rectified using 12 Ground Control Points (GCPs) surveyed using the Differential GPS (DGPS) technique. The LIDAR-based DEM is then used as a source for other GCPs that are used to orthorectify the Ikonos image. Different 2-D transformation models are used with different sets and distributions of GCPs. The planimetric Root Mean Square Errors (RMSE) achieved, using these models, is about 5 m. The relationship between the planimetric errors and the elevations of the checkpoints suggested using 3-D to 2-D transformation models. Three different types of these models are examined and their results are reported. The results showed that less than 2 m mapping accuracy could be achieved using the 3-D to 2-D transformation models. These results suggest that using a single panchromatic Ikonos image together with a 1-m resolution LIDAR-based DEM could achieve the required planimetric accuracy for 1:5000 topographic maps.