tanbul 2004
ith of ca. 3
al areas, by
he existing
lasses. The
' for giving
l-in gaps in
rmation in
it to the
m of these
te network
'onsist of a
System has
eby the 3D
rmation on
ve certain
nain being
uring their
re sensor
input data
)perational
on of the
sults with
this work
s well as
centage of
| extracted
cted roads
998). The
icy of the
ssessed by
cted roads
| for the
esented in
and-alone
hoimages
thoimage
1 imports
data (e.g.
| as the
are saved
ted into
? system,
>rsion for
for roads
)vements
OMIRO,
N
resented
Geneva.
t the test
cover in
is the
jS are in
types in
n of the
] fis. 1
ger and
) with a
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B2. Istanbul 2004
In Thun, the colour orthoimages were produced by swisstopo
from aerial images of scale 1:16,000 using the DHM25. The 50-
cm orthoimage is part of the nationwide dataset Swissimage
(produced from 1:30,000 imagery with 15 cm lens) with a
planimetric accuracy of about 1 m. The images for 20 cm and
60 cm were taken in spring 2003, and for 50 cm in summer
1998. An orthoimage created from ADS40 summer images
using the DHM25 is also available. Due to weaknesses in the
control point distribution and the bundle adjustment of the
ADS40 images, a discrepancy between the ADS40 orthoimage
and the 20cm orthoimage has been observed. A non-exhaustive
comparison with manually selected feature points shows that
the discrepancy varies between 0 and 80 cm. However, smaller
differences also exist between the 1998 and 2003 aerial film
orthoimages, caused by errors in the sensor orientation. Thus,
the real accuracy of road extraction in image space is higher
than the accuracy values derived from comparison between
datasets (incl. the reference data), which have varying
orientation errors.
Thun Geneva
Area (sq. km) 2.66 * 2.68 4.38 * 3.0
Height range (m) | 560 ~ 2200 375 — 1200
Open rural, Open rural,
Villages, Several villages,
Landscape Many small Forest,
settlements Large fields with
bare soil
aerial film (50cm)
IKONOS PSM
(1m)
Quickbird PSM
(70cm)
aerial film
(20cm, 50cm, 60cm)
Imagery type
(orthoimage
pixel size)
ADS40 (30cm)
Table 1. Test site description and image specifications.
Figure 1. Overview of test sites: (a) Thun, (b) Geneva.
The Geneva test site (Fig. 1b) is near the city of Geneva,
containing several larger villages, forest and a river. Another
difference to the Thun site is that the scene contains grasslands
and large fields of bare soil. In addition, many road-like lines
are observed in the fields. The aerial orthoimage came from the
74
Swissimage dataset. IKONOS and Quickbird images were also
acquired in May 2001 and July 2003 respectively. The
pansharpened (PSM) orthoimages of IKONOS and Quickbird
were produced by a software system developed at IGP using a
2-m grid laser DTM with 0.5-m accuracy and had a planimetric
accuracy of 0.5-0.8 m, estimated using 20-50 check points
measured in 25-cm orthoimages of the Canton Geneva,
produced using the same laser DTM (however with 1m grid
spacing) and with ca. 0.5-m planimetric accuracy).
The reference data for the Thun and Geneva test sites were
measured manually, by swisstopo in 20-cm pixel size aerial
orthoimages and by ETH Zurich in the Swissimage
orthoimages, respectively. The tests were performed on a DELL
PC with Pentium 4, !.8GHz CPU and 1GB RAM running
Windows XP.
4. RESULTS AND DISCUSSION
4.1 Thun Site
Completeness and correctness is sufficient for all images in the
Thun site, with slightly inferior results for the 50-cm and 60-cm
pixel size orthoimages. Although the pixel size of the ADS40
image is slightly more than the 20 cm of the aerial film
orthoimage, the results achieved are almost identical. Typical
results of road reconstruction and junction generation are
presented in Figs. 2-5, where the VEC25 and the extracted
roads are shown as white and black lines. In each figure, (a),
(c). (d) are the orthoimages with pixel size 20 cm, 50 cm and 60
cm respectively, while the 30-cm pixel size ADS40 orthoimage
is shown in (b).
Fig. 2 is a scene with a four-road junction. Road surface and
road sides are clear except at the left side of the figures, where a
tree occludes the road. The scenes in Fig. 3 and Fig. 4 are
slightly complex compared with that in Fig. 2. More shadows
and occlusions are observed. In the settlement areas, some road
sides are not defined. In Fig. 5, a first-class road is connected
with two third-class roads at two junctions. The roadmarks on
the first-class road are visible in all images, but are weak in the
lower resolution images. The examples show that roads are
generally correctly extracted from all images. Road junctions
are also well formed. This observation is confirmed by the
external evaluation of the extraction results using the reference
data (see Table 2). To account for the discrepancy between the
ADS40 and aerial film orthoimages, the buffer distance was set
to 2 m, when assessing the results from the ADS40 orthoimage.
Figure 2. Examples of road extraction and junction generation
in scenes with well defined roads.
3
