Ground truth extraction from LiDAR data for image orthorectification

Abstract

The availability of high accuracy GCPs (ground control points) and DEMs (digital elevation models) becomes the key issue for successful implementation of an image orthorectification project. It is a very difficult task for collecting a large number of high quality GCPs by using traditional methods to meet all the requirements for digital photogrammetric and orthorectification process. Airborne light detection and ranging (LiDAR) - also referred to as airborne laser scanning (ALS), provides an alternative for high-density and high-accuracy three-dimensional terrain point data acquisition. One of the appealing features in the LiDAR output is the direct availability of three dimensional coordinates of points and intensity data in object space. With LiDAR data, high- accuracy and high-resolution intensity image, hillshade DSM (digital surface model) image, and DEM can be generated. Due to high planimetric accuracy characteristics of LiDAR data, ground truth can be extracted from these LiDAR-derived products (e.g., hillshade image and intensity image). This study investigated the feasibility of using LiDAR-derived hillshade DSM image and intensity image to extract ground truth for aerial image orthorectification. Two sets of GCPs were extracted from hillshade image and intensity image separately, and then were used as the inputs for aerial triangulation processing. LiDAR- derived DEM was then employed for differential rectification to produce the final orthoimage. The assessment of the planimetric accuracy of orthorectified images by using different set of GCPs was conducted by comparing the coordinates of some checking points from orthoimages and correspondent GPS surveyed coordinates.