The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part BI. Beijing 2008 
242 
2. DATA COLLECTION, DATA CHARACTERS AND 
BASIC SOFTWARE 
The experiment data were collected with Leica HDS3000 on the 
campus of NUIM (National University of Ireland, Maynooth). 
The target is John Hume Building. The equipments are shown 
in Fig.l. 
Fig.l field and raw LiDAR data collected 
The raw LiDAR data showed in Fig.l was visualized by Leica 
cyclone5.8, which provides integrated LiDAR data processing 
environment. This image is one scanworld of the whole survey 
data. Each scanworld includes the LiDAR data acquired from 
one spot. The data format is *.imp. Cyclone doesn’t provide 
functions of automatic linear feature extraction and edge 
detection, though it has manual edge detection function. The 
raw LiDAR data have to transform into other formats, which 
could be read by other software, for realizing edge detection and 
linear feature extraction. Fortunately, cyclone can export 
scanworld into several formats, which are widely used, such as 
DXF, xyz and txt. Different format contains the same X, Y, Z 
coordinates of base spot and those of objects. However, the 
points are recorded in various orders in different formats. For 
example, Fig.2 shows the difference order. 
if 
Xttï) «*® M® «ft® 
|-97.182 022,-1*5.022934,-a.312119 
-32.723495,-14.977188,-1.460953 
-32.117508,-14.925095,-1.472153 
-21.609055,-9.919479,-1.49437 
-52.082962,-24.931412,-1.506363 
-45.725479,-24.833939,-1.410263 
-32.118851,-14.963242,-1.472916 
-29.566116,-16.769363,-1.413956 
-21.536728,-9.987167,-1.491806 
-13.990128,-8.030991,-1.594864 
_ M 
U> l, Co: 
mi® «s® as® «ft® 
-97.182022 
-154.578262 
-154.695328 
-154.579239 
-154.692337 
-133.154739 
-134.661636 
-121.017593 
-77.537216 
45.022934 -0.312119 
-70.883469 -0.494980 
-70.757736 -0.495529 
-70.883987 -0.331833 
-70.756363 -0.332993 
-70.754349 -0.168259 
-67.546188 -0.146988 
-65.881973 -0.293503 
-59.207748 -0.132889 
40.565567 -0.425613 
Ln 1, Co! 
Fig.2 difference order of coordinates in various formats 
The first records in the two files are the same. They are the 
value of base point. This show the points are recorded without 
definite order. This causes some difficulties in data processing. 
Lidar data are also different from raster image. They are vectors. 
And they have real 3D information. As to raster image, 
parameter Z stands for grey value of a grid. This means raster 
image is not true 3D. Most tradition edge detection algorithms 
for raster image are usually not suitable for Lidar data. 
3. ALGORITHM FOR EDGE DETECTION OF 
TERRESTRIAL LIDAR DATA 
The key step to realize linear feature extraction is edge detection. 
Since terrestrial LiDAR data has different characters from raster 
image and airborne LiDAR data, it is necessary to build 
effective and efficient algorithms to realize edge detection. 
Some researches have put forward results of traditional 
algorithms in edge detecting of LiDAR data, such as Sobel, 
Kirsch, LOG, Canny and Roberts operators (Li Qi, 2003; .Lai 
Xudong, 2005). These algorithms were demonstrated not 
suitable for edge detection of LiDAR data. Zheng (2005) used 
fractal dimension methods to distinguish road from river in 
IKONOS image. Studies show that the nature background in the 
image is accorded with Brown movement model, which shows 
self-comparability of its local gray, some relativity with less 
variety of grey-level between neighbor pixel. On the contrast, 
the man-made objects have the brims with grey breaks and 
relative grey varieties between neighbor brim pixels (Yang, 
2003). With the slide window, in nature background, its fractal 
dimension is small. If more brims of man-made objects, its 
fractal dimension is bigger. Since the nature and man-made 
objects have different fractal dimensions generally, Zheng 
(2005) demonstrated fractal dimension theory can be used 
distinguish road from river element from remote sensing images 
automatically. The research gave illumination to edge detecting 
of terrestrial LiDAR data. If there is an edge, fractal dimension 
of the grid that it belongs to must has bigger value than that of 
inner grid. Since LiDAR data the different features, the 
algorithm used in Zheng’s research has possibility to be used for 
edge detection of terrestrial LiDAR data. 
In general, fractal dimension theory mainly has three algorithms, 
line-divider method, slop-direction method and Triangular 
method. This research chose the third one for its advantages. 
Triangular method is same in nature as the slop direction one. 
The fractal dimension can be obtained by calculating the 3D 
surface area of the remote sensing image. However, the 
algorithm is more precise. The study shows: the nature 
background is accorded with fraction Brown movement model, 
which is one of the classical models in fractal signals (or image) 
study for its excellent characters. By the way, the surface area is 
measured by the following equations (Zheng, 2005): 
log^ = C + 51ogG (l) 
D = 2-B (2) 
Parameter A stands for the measured surface area, G is the step, 
B is the slope, C is a constant, and D is the fractal dimension. 
Due to no entire self-similarity, it often shows the similarity 
from the statistic extent. In practice, a series of G and A values 
are selected in the logarithm coordinate system and are fit with