1255
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Voi. XXXVII. Part B4. Beijing 2008
LiDAR technology is one kind of high technology integration
system, which can explore objects and measure distance by the
way of laser scanning. With this system we can obtain urban
data surface model easily. How to build 3D models of digital
city is a big issue to discuss by professional people. Working
with point clouds of LiDAR data and images is good method,
which only need point clouds. So, we build over 1000 buildings
and some bridges with LiDAR data in Nanjing. Following
figure 5 showed footprints and three-dimensional models.
Figure 5 LiDAR footprints display on three-dimensional
models
We can make three-dimensional models according points and
images. The cloud points can make more perfect description on
proof of building than others. We also analyze accuracy of them.
We found that the error of elevation is smaller than 0.5 meters
and it can describe surface of top of buildings.
3.4 More research work we did
Another part of research work is field survey with GPS and
Total Stations. We want to know how much deviation of
LiDAR points in DEM. We selected 4 lines with 0.6KA/L
1.0KA/L 0.7KM and 1.1 KM long, and they are on the asphalt of
the road surface. The footprints are all on break line, so we can
distinguish them between others. On the other hand, we got the
coordination of these points with GPS, and then contrasted with
them. We also survey 45 transects of the road with GPS, and
made other contrast to find how much differences they had.
Figure 6 break lines survey in two ways
This figure 6 is divided into three parts: left, middle, and right.
The left show the all concerned site and the middle is the
details in black circle in the left. The right is same site as the
middle but in shadowed feature of LiDAR data. They have
same points that were connected by light lines and showing
with small black or white circles. From our research it can be
found that we can obtain break lines with LiDAR data, but it is
dependent on the density of footprints. In other words, if the
mean distance of two LiDAR points is bigger than one meter,
acquiring ordinary break lines will be more difficult. From
comparison between 45 transects of the road with GPS and
obtained with LiDAR data, we conclude that there is no much
different in them, and we can do transects of the road with
LiDAR systems instead of field survey.
4. CONCLUSIONS
From our research work, we have got followed conclusion on
accuracy of LiDAR data:
1, It is important that LiDAR data acquiring. Stabilization of
flight is also important. It is the most important factor to affect
accuracy of post processing.
2, Accuracy of DEM is very good, mean error is smaller than
0.15 meters in elevation value. It can be used in design projects
of transportation and all kinds of survey projects.
3, Accuracy of DOM is good, mean error is smaller than 1
meter planar value. It can be used in almost survey projects and
3 dimensional model building projects.
4, Accuracy of DLG is OK, mean error of plane smaller than 1
meter. It can be used in drawing under 1:2000 scales
topographic.
5, it is one of the best ways to found 3D models of digital city
with data of LiDAR system.
REFERENCES
Arttu Soininen, 1998-2004. Manual for four modules of
Terrasolid software. Terrasolid Inc, Finland.
Gary C. Guenther, A. Grant Cunngham, Paul E. Larocque, and
David J., 2000. The accuracy challenge in airborne LiDAR
bathymetry. Proceedings of EARSel-SIG-workshop LiDAR,
Dresden/FRG. pp 16-17.
Han Wenquan, et al. 2007, Analysis of workflow of
producing DLG with LiDAR technology, bulletin of surveying
and mapping, No.5 (sum. 362), pp58-60.
Lukas Wotruba, et al. 2005, assessment of sensor
characteristics of an airborne laser scanner using geometric
reference targets, ISPRS WG III/3, III/4, V/3 Workshop "Laser
scanning 2005", Enschede, the Netherlands, September 12-14.
Wang Xiaohui, et al. 2007, precision tests LiDAR data products.
Engineering of surveying and mapping. pp67-69.
Zhang dong, 2005, Wu Han University. Based on the LIDAR
data and aerial images of three-dimensional reconstruction of
urban housing. pp8-17.
ACKNOWLEDGEMENTS
This work was supported in part by National 863 plans projects:
software platform of processing LiDAR data. Project No. is
2006AA12Z101.