ON THE MATCHING ACCURACY OF RASTERISED SCANNING LASER ALTIMETER DATA

Abstract

For certain applications irregularly distributed scanning laser altimeter data need to be rasterised - such as for use in GIS systems and for creating DEMs. Also, least squares matching on a raster grid can enable the measurement of planimetric and height shifts between overlapping strips of laser data. The shifts are a manifestation of errors in the laser altimeter, most of which are caused by the positioning elements of the system (GPS and/or INS). These translations form the input into a block adjustment to correct for relative and absolute errors. Here a discussion of the issues related to deriving a regular grid of 2.5D points from the original data is presented, with particular reference to the interpolation method, grid size, and quantisation level. An interpolation method based on a TIN of the original points with a grid size that relates as closely as possible to the point density at acquisition is found to give the best results. 8-bit quantisation is found to be sufficient for height differences of up to 100m.