Erik de Min
ON THE ADJUSTMENT OF OVERLAPPING STRIPS OF LASERALTIMETER HEIGHT DATA
M.J.E. Crombaghs, R. Brügelmann, E.J. de Min
Ministry of Transport, Public Works and Water Management, The Netherlands
Survey Department
m.j.e.crombaghs @mdi.rws.minvenw.nl
KEY WORDS: Laser, accuracy, corrections, quality control, data processing, GPS, INS, software
ABSTRACT
From the beginning of the nineties up until now the Dutch Survey Department (Meetkundige Dienst Rijkswaterstaat) has
put out to contract several airborne laser altimetry projects. In order to be able to check and to improve the quality of the
delivered laser data, a strip adjustment procedure concerning the heights has been developed at the Survey Department.
This enables to correct for systematic height errors occuring per strip, that is for an offset as well as an along-track and
an across-track tilt of every strip. Unfortunately, this way of considering the strip as stiff and flat board turned out not
always to suffice. Therefore, further correction tools have been created to eliminate possible occuring cross strip
parabolic effects and to handle strip torsions or periodic effects. Apart from quality analysis of real data, some
simulation computations have been done to investigate the maximal achievable theoretical precision in case of adding
much more ground control points and extra cross laser strips. Averaging the heights of an area of 0.01 km”, a precision
of 1-2 cm seems to be achievable for the mean height.
1 ‘RIBBELS’ IN LASERALTIMETER HEIGHT DATA
For a few years, a new digital elevation model for the surface of the entire Netherlands has been acquired using airborne
laser scanning with a density of one point per 16 n?. The new DEM is expected to be completed in 2001. For the
acquisition task, the Survey Department contracted several companies. The flightplans for the acquisition of laserdata
resemble those of photogrammetry. The terrain is being sampled in blocks consisting of several overlapping parallel
strips. These strips can be tens of kilometres in length and several hundred metres wide. Besides, a certain number of
cross strips is flown for checking purposes and some ground control ‘points’ per block are measured.
Figure 1a: *Ribbel' in heightdata marked by dots Figure 1b: Point density grid (the darker the denser)
(the straight line is a real ditch; white pixels — no data)
The Survey Department has developed a quality control procedure for laser data. Part of this procedure is a visual
inspection of the data. A striking property is the frequent presence of erroneous line structures in ‘hillshades’ created
from the laser data (see figure 1a). Inspection of these so called ‘ribbels’ (ridges) showed that they occur in strip
overlaps. Figure 1b shows the density of laser altimetry points. The density is about doubled (dark gray) in areas where
two strips overlap. The ribbels are caused by height differences between laserpoints of two neighbouring strips. These
height differences are caused by offsets and tilts of the strips.
230 International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B3. Amsterdam 2000.
