REVISION AND RECONSTRUCTION OF 3D BUILDING DATA BY INTEGRATING
STARIMAGER/TLS IMAGERY AND COMPLEMENTARY DATA
Masafumi NAKAGAWA*, Ryosuke SHIBASAKI**
*Graduate School of Frontier Sciences, Institute of Environmental Studies
mnaka(@iis.u-tokyo.ac.jp
and
**Center for Spatial Information Science
University of Tokyo
4-6-1 Komaba, Meguro-ku, Tokyo, 153-8505
shiba@skl.iis.u-tokyo.ac.jp
KEY WORDS: Digital Photogrammetry, TLS(Three Line Sensor), 3D Mapping, Change detection, Data revision
ABSTRACT:
Change extraction of building is needed to revise building data effectively. Many change detection algorithms use
height difference analysis using temporal data such as LIDAR. However, remarkable changes of buildings cannot be
detected in urban dense areas. On the other hand, building textures might change with higher possibility. However, the
change cannot be detected due to influences of shadows and occlusion caused by nearby buildings in urban dense areas.
Therefore, we have proposed a method to revise 3D building data by integrating texture change (roofs and walls) and
3D shape change of buildings using STARIMAGER/TLS (Three Line Sensor).
1. INTRODUCTION
Change extraction of building is needed to revise
building data effectively. The change may be “shape
change” of roofs and walls, “texture change” of roofs and
walls or “attribute change” such as owner’s name. These
changes occur due to reconstruction or demolition of the
building.
Many change detection algorithms use height difference
analysis using temporal data such as LIDAR. However,
remarkable changes of buildings cannot be detected in
urban dense areas since most of the buildings are built on
full lot size due to very small lot area. Therefore, when
only building shapes are used for building change
detection, a correctness of the change detection in dense
area is lower than that in suburbs in many cases.
On the other hand, building textures might change with
higher possibility. However, the change cannot be
detected due to influences of shadows and occlusion
caused by nearby buildings in urban dense areas. Though
digital aerial photos and satellite images are used in
existing research, these images have disadvantages such
as occlusion and low resolution. Moreover, when only
roof textures are used, all changes cannot be detected in
images even if buildings change.
Therefore, we have proposed a new method to revise 3D
building data automatically by integrating texture
changing (roofs and walls) and 3D shape changing of
buildings using STARIMAGER/TLS (Three Line
Sensor) in this paper.
2. METHODOLOGY
At first, 3D building texture data, which is assumed as
existing data, are prepared. The data include not only
roof textures but also wall textures. If the data do not
exist, they can be generated from TLS images, which are
acquired at different time period than the TLS data for
change detection.
Next, candidates of building shape changes are detected
by using DSM of existing 3D building data and DSM
generated from TLS image acquired for change detection.
Approximate changes, which are probability of shape
change in urban areas, are detected. This preliminary
information is assigned to each building.
Then, existing 3D data are projected in TLS images.
Building changes are detected by using the preliminary
information and changes of textures in the TLS images.
Not only roofs but also walls are referred in this
processing. Building shadows are extracted from TLS
images by using the temporal information of the data
acquisition. TLS images are enhanced not to influence
the texture change detection. .
Moreover, when buildings with changes are detected in
the existing TLS images, new 3D data are reconstructed
at the same location by using initial values. The initial
values are based on DSM generated from TLS images, a
building template model and surrounding information.
1324
Interi
3.
STA
sensi
CCD
of ez
same
an ai
the g
grour
TIS
3cm)
the s
advar
infort
3D bt
data, €
roof t
exist, |
acquir
change
Additi
require
metho
overla:
figure
corres]
Theref:
Figure.
