ArticlePDF Available

Abstract

This paper investigates automatic prediction of seismic building structural types described by the Global Earthquake Model (GEM) taxonomy, by combining remote sensing, cadastral and inspection data in a supervised machine learning approach. Our focus lies on the extraction of detailed geometric information from a point cloud gained by aerial laser scanning. To describe the geometric shape of a building we apply Shape-DNA, a spectral shape descriptor based on the eigenvalues of the Laplace-Beltrami operator. In a first experiment on synthetically generated building stock we succeed in predicting the roof type of different buildings with accuracies above 80%, only relying on the Shape-DNA. The roof type of a building thereby serves as an example of a relevant feature for predicting GEM attributes, which cannot easily be identified and described by using traditional methods for shape analysis of buildings. Further research is necessary in order to explore the usability of Shape-DNA on real building data. In a second experiment we use real-world data of buildings located in the Groningen region in the Netherlands. Here we can automatically predict six GEM attributes, such as the type of lateral load resisting system, with accuracies above 75% only by taking a buildings footprint area and year of construction into account.
SHAPE BASED CLASSIFICATION OF SEISMIC BUILDING STRUCTURAL TYPES
R. Sulzer1,
, P. Nourian1, M. Palmieri2, J.C. van Gemert1
1Delft University of Technology, Netherlands - (P.Nourian, J.C.vanGemert)@tudelft.nl
2Arup Amsterdam, Netherlands - Michele.Palmieri@arup.com
KEY WORDS: Seismic Building Structural Type, Classification, Machine Learning, Point Cloud, Shape Descriptor, Shape-DNA
ABSTRACT:
This paper investigates automatic prediction of seismic building structural types described by the Global Earthquake Model (GEM)
taxonomy, by combining remote sensing, cadastral and inspection data in a supervised machine learning approach. Our focus lies on
the extraction of detailed geometric information from a point cloud gained by aerial laser scanning. To describe the geometric shape
of a building we apply Shape-DNA, a spectral shape descriptor based on the eigenvalues of the Laplace-Beltrami operator. In a first
experiment on synthetically generated building stock we succeed in predicting the roof type of different buildings with accuracies above
80%, only relying on the Shape-DNA. The roof type of a building thereby serves as an example of a relevant feature for predicting
GEM attributes, which cannot easily be identified and described by using traditional methods for shape analysis of buildings. Further
research is necessary in order to explore the usability of Shape-DNA on real building data. In a second experiment we use real-world
data of buildings located in the Groningen region in the Netherlands. Here we can automatically predict six GEM attributes, such as
the type of lateral load resisting system, with accuracies above 75% only by taking a buildings footprint area and year of construction
into account.
1. INTRODUCTION
Knowledge about the seismic vulnerability of existing building
stock is of vital importance in seismic risk management, e.g. for
the design and development of seismic retrofit strategies. The
seismic building structural type (SBST) reflects the main load-
bearing structure of a building and therefore its behaviour under
seismic load (Geiß et al., 2015). However, for numerous areas in
earthquake prone regions this information is often outdated, un-
available, or simply not existent (Geiß, 2015). Traditional meth-
ods to gather this information, such as building-by-building in-
spections, are costly and highly time-consuming, making them
unfeasible for assessing large building inventory. For this reason
the use of remote sensing data and ancillary geo-information has
been proposed to allow a fast acquisition of SBST information on
urban and regional scale. Subsequently, machine learning algo-
rithms may be used to analyse the gathered information, e.g. to
classify a building stock into groups with similar SBSTs (Borzi
et al., 2011, Christodoulou et al., 2017, Geiß et al., 2015, Lugari,
2014, Pittore and Wieland, 2013, Sarabandi, 2007). However, ex-
isting approaches for such a workflow often deliver highly aggre-
gated results in terms of their spatial or typological granularity,
and therefore prevent a precise seismic assessment.
The GEM foundation1provides an internationally standardised
scheme for seismic risk assessment. Within this scheme, the
GEM building taxonomy was developed, to allow the uniform
classification of buildings with regard to their expected seismic
behaviour. To this end, the GEM taxonomy describes a building
with 13 attributes that uniquely determine its SBST, such as the
lateral load resisting system, the floor or the exterior wall type.
In this research we aim to predict GEM attributes for buildings
located in the Groningen region in the Netherlands. In this region
seismic risk is induced by the extraction of gas from the large
Corresponding author at raphaelsulzer@gmx.de
1https://www.globalquakemodel.org/
Groningen gas field (Muntendam-Bos et al., 2015). This leads to
a unique situation in which the traditional, mainly non-resilient
building stock of the Groningen region is exposed to recurring
seismic events with minor intensities. Arup Amsterdam is com-
missioned with a large scale seismic assessment of the region.
Based on existing building information, such as building plans
and a small number of in situ building inspections, Arup has de-
veloped a method to determine GEM attributes for around 10% of
the 250.000 affected buildings. We use this data as a training set
in a supervised machine learning approach, by combining it with
openly available geo-data. For now, we investigate into predictive
accuracies on the training set. In the future, we want to predict
GEM attributes for the entire building stock of the Groningen re-
gion.
Previous research in this field often concludes that shape features
are important properties when trying to infer seismic building
structural types (SBSTs). An example are pitched roofs with
large spans that can be an indication for frame-based structures
(Nederlandse Aardolie Maatschappij BV, 2015). However, ex-
tracting such properties from unstructured remote sensing data is
often not trivial. To this end we explore a new method to de-
scribe the shape of buildings by using Shape-DNA. Shape-DNA
represents the global shape of an object with the normalized be-
ginning sequence of the eigenvalues of the Laplace-Beltrami op-
erator (LBO). By using Shape-DNA to describe the shape of a
building we expect to improve the prediction of SBSTs.
We make the following contributions. We start in a controlled
environmnet by generating synthetic building models in the form
of polygonal meshes with different mesh densities, three differ-
ent roof types and different extensions and classify them only
by using Shape-DNA. Then, for the real-world Groningen SBST
prediction, we first extract building features, such as the foot-
print area and year of construction, from a cadastral dataset. We
generate 3D building models in the form of polygonal meshes
by using footprint polygons and an aerial laser scanning (ALS)
point cloud of the Groningen region. From the building meshes
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLII-4/W10, 2018
13th 3D GeoInfo Conference, 1–2 October 2018, Delft, The Netherlands
This contribution has been peer-reviewed.
https://doi.org/10.5194/isprs-archives-XLII-4-W10-179-2018 | © Authors 2018. CC BY 4.0 License.
179
we extract Shape-DNA and other geometric features. We then
represent each building by a feature vector. Supervised machine
learning techniques, such as Random Forests (RF) and Support
Vector Machines (SVM) are used to learn patterns in the training
set based on the feature representations and estimate the GEM at-
tributes of buildings. We also assess the influence of each of our
features for the SBST prediction, based on predictive accuracies
and feature importance scores.
2. RELATED WORK
Automatic prediction of SBSTs for detailed pre-event seismic
vulnerability assessment has only been attempted in a few, largely
heterogeneous approaches (Borzi et al., 2011, Christodoulou et
al., 2017, Geiß, 2015, Lugari, 2014, Pittore and Wieland, 2013,
Sarabandi, 2007). The works differ in their input data, from
terrestrial images (Pittore and Wieland, 2013) to aerial (Lugari,
2014) and satellite imagery (Borzi et al., 2011, Geiß, 2015, Lu-
gari, 2014, Pittore and Wieland, 2013, Sarabandi, 2007), aerial
laser scanning point clouds (Christodoulou et al., 2017, Geiß,
2015, Lugari, 2014) or a combination of the aforementioned.
They extract topologic (Geiß, 2015, Sarabandi, 2007), geomet-
ric (Borzi et al., 2011, Christodoulou et al., 2017, Geiß, 2015,
Lugari, 2014, Pittore and Wieland, 2013, Sarabandi, 2007), vi-
sual (or spectral) (Borzi et al., 2011, Geiß, 2015, Lugari, 2014,
Pittore and Wieland, 2013), geographic and semantic features
on building and building block level (Geiß et al., 2015, Wieland
et al., 2012). Furthermore, different definitions of SBSTs, e.g.
stemming from the European Macroseismic Scale (EMS) are pre-
dicted. In this paper, we build on top of the ideas of these works.
However, we predict SBSTs defined by the GEM building tax-
onomy, which describes a building in significantly more detail
than previously used typologies. Additionally, we investigate into
improving the commonly used geometric features by applying a
spectral shape descriptor.
Geometric shape features Geometric features are often regarded
as the most important proxies to determine SBSTs. Simple fea-
tures, such as the footprint area (Borfecchia et al., 2010, Lugari,
2014) or the building height (Pittore and Wieland, 2013) can
already be useful to distinguish smaller from larger structures.
However, to predict a more detailed description of SBSTs, a more
detailed description of the building geometry may be necessary.
Examples of features that describe the building geometry are the
shape of the footprint or the roof type of a building. The shape of
the footprint can e.g. be described by its slenderness, convexity
and irregularity (Sarabandi, 2007) or compactness, elongation,
density and many other parameters (Geiß et al., 2015). Most of
these features are calculated by using some sort of quotient of the
area, perimeter, longest axis or minimum bounding rectangle of
the footprint. A way to include information about the roof of a
building is to first classify the roof shape into different types such
as flat simple, flat multi-level, pitched and complex (Borfecchia
et al., 2010, Lugari, 2014) or flat, low slope and steep slope (Sara-
bandi, 2007) and then use this information as a categorical feature
in the SBST classification. In our approach we do not want to first
classify the roof structures into different typologies. Instead we
extract shape features from a complete 3D representation of the
building and directly included them in the SBST classification.
We do this for a variety of reasons. Extracting local shape fea-
tures, such as the roof angle, from a semantically unstructured
geometric representation of a building is not trivial. For complex
shapes, extensions on the building of secondary importance can
significantly influence the values of such features (see e.g. Fig-
ure 4). Moreover, it is not obvious which local properties are
generally relevant for the classification of a variety of different
SBSTs. Even though features, such as the roof steepness, may be
suited to distinguish lateral load resisting systems, such as wall
based systems from frame based systems, they lack discrimina-
tive power to distinguish them from other types. Other relevant
features may be subject to common building practice or rules and
regulations in a particular area or construction period. Thus, find-
ing relevant shape features for an entire building stock can be a
cumbersome task, even for an expert in the field. On the other
hand, global building properties that are easier to obtain, such as
the volume or the surface area of a building might often not be
sufficient for describing complex building geometries. It is there-
fore desirable to describe the geometric shape of a building in a
way that all relevant building features can be captured and ex-
tracted. To achieve this we make use of a global shape descriptor
called Shape-DNA.
Visual features Several of the aforementioned studies (Borzi et
al., 2011, Geiß, 2015, Lugari, 2014, Pittore and Wieland, 2013)
use visual information extracted from satellite, aerial or terres-
trial images. However, they often come to the conclusion that
visual information is only of secondary importance compared to
shape features. Furthermore, visual information can sometimes
be misleading, e.g. the facade of a building does often not allow
conclusions about its structure (Christodoulou et al., 2017). In
this paper we do not include visual information in the classifica-
tion, however, we do believe that it is worth to investigate into its
use for SBST classification in the future, especially considering
the vast amount of sources for this type of data.
Semantic features Semantic features may be the building age,
which is often estimated using multitemporal images, or the
buildings’ occupancy type (residential, commercial, industrial),
that can e.g. be gained from tax assessors data (Sarabandi, 2007).
In this work we make use of a cadastral dataset that includes both,
the year of construction and occupancy type. However, we only
use the year of construction as a feature in our classification and
not the occupancy type (see section 3.4). The year of construc-
tion can help to identify common building practice or rules and
regulations applied at different construction eras.
Geographic features In our approach we only use building-level
information for classification. One exception is a binary feature
describing the presence of a directly adjacent building (see sec-
tion 3.4). Generally, the building geometry allows characterising
individual buildings, whereas information on building block level
characterises the geographic setting in which the respective build-
ings are embedded in (Geiß et al., 2015). This can be valuable
information in many stages of the process. In the classification
process it can help to prevent misclassifications due to unlikely
geometries of single structures. It can provide information on
different spatial scales, also with regard to data collection. This
should be addressed in future implementations.
Classifiers To predict SBSTs often a supervised learning ap-
proach is used, that takes in a training set of buildings represented
by a vector containing (some of) the aforementioned features and
labeled with their corresponding SBSTs. Based on the feature
representation, classifiers such as a Random Forest (RF), a Super
Vector Machine (SVM) (Geiß, 2015) or an Artificial Neural Net-
work (Lugari, 2014) can find patterns in the labeled training set
and predict the class of unknown buildings. One downside of the
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLII-4/W10, 2018
13th 3D GeoInfo Conference, 1–2 October 2018, Delft, The Netherlands
This contribution has been peer-reviewed.
https://doi.org/10.5194/isprs-archives-XLII-4-W10-179-2018 | © Authors 2018. CC BY 4.0 License.
180
latter classifier can be that it requires a larger amount of training
samples, thus we make use of RFs and SVMs in our approach.
Seismic structural building type Next to different input data,
extracted features and classifiers, a large source of variation in
previous works is reflected in the predicted typologies. While
some works use their own definitions of building typologies
(Borzi et al., 2011, Geiß et al., 2015, Lugari, 2014), others pre-
dict typologies defined in HAZUS-MH2(Sarabandi, 2007), the
World Housing Encyclopedia3or the European Macroseismic
Scale4(Wieland et al., 2012). These typologies are often too
broad (EMS), are not primarily designed for SBST classifica-
tion (WHE) or focus on specific regions (HAZUS) (Brzev et al.,
2013). In our approach we predict GEM typologies as they are
internationally standardised and were developed to allow the uni-
form classification of buildings with regard to their expected seis-
mic behaviour. The GEM taxonomy describes a building with
13 attributes that uniquely determine its SBST. For a detailed
overview of all attributes we refer the reader to the GEM Building
Taxonomy (Brzev et al., 2013). It is, however, important to note
that the possibility to predicted a certain typology is constrained
by the available training data. In a supervised learning approach
it is only possible to predict typologies that are present in the
learning sample. The generation of our training sample required
in-depth seismic engineering knowledge and the availability of a
variety of datasets (see section 3.4). Acquiring such a training
sample can be challenging in many cases.
3. METHOD
3.1 Generation of building models
3.1.1 Synthetic building models In this section we report on
the generation of synthetic building models from piecewise pla-
nar segments. We generate building models with flat, pitched and
gabled roofs and apply different mesh densities. To this end, we
randomly draw the footprint length las 4m < l < 8mand width
was 5m < w < 9m, the building height has 3m < h < 12m,
as well as the roof angle αas 15< α < 60for buildings with
pitched and gabled roofs. Additionally we create models with a
cubic extension with l, w, h ∈ {1m, 2m, 3m}. This results in
a dataset with in total 1200 building models: 100 buildings with
different dimensions ×3 roof types ×4 extension types.
Next we sample points of each planar building segment in a
20cm ×20cm grid. We apply a Delaunay triangulation to
each point segment and glue the resulting meshes together to get
one polygonal mesh for each building model. For an additional
dataset, we generate one mesh for each roof type where points are
sampled in a 50cm ×50cm grid. To resemble the meshing we
achieve for the building models generated from remote sensing
data we also generated one mesh without vertices on the vertical
walls (cf. Figure 2e). Additionally, we remesh these models and
apply Laplacian smoothing (see also section 3.1.2) to achieve a
better mesh quality for the models without wall points. This last
dataset will be used to assess the influence of the meshing on the
Shape-DNA (section 4.1).
3.1.2 Building models from remote sensing and cadastral
data While this step is not strictly necessary for SBST clas-
sification it can facilitate the extraction of 3D building informa-
tion. As an example, a 3D model of the building constructed with
2HAZUS - https://www.fema.gov/hazus
3World Housing Encyclopedia - http://www.world-housing.net
4European Macroseismic Scale - https://www.gfz-potsdam.de/ems98
piecewise (planar) segments can help to extract local shape fea-
tures, such as the number of roof segments or the facade area.
However, even though 3D building reconstruction from remote
sensing data has been an active area of research for the past 20
years, it is still a cumbersome task to reconstruct urban scenes
with buildings of arbitrary shape (Haala and Kada, 2010). We
do make use of one tool that is frequently applied in 3D build-
ing reconstruction: we apply a region growing segmentation to
the ALS point cloud of the building roof to identify different roof
segments (Grilli et al., 2017). We then use the number of roof seg-
ments and the average of all segment angles as features (see also
section 3.3.2). Since further progress in reconstruction would re-
quire considerable effort, we want to, at least for now, avoid this
topic. Instead, we apply a Delaunay triangulation to the ALS
point cloud and the footprint polygon and merge them to gain
a 3D polygonal mesh that allows the extraction of Shape-DNA.
The full process is depicted in the flowchart in figure 1.
ALSpoint cloud
of the area
Delaunay
triangulation
building
point clouds
extract naked
vertices
point-
in-polygon
test
inside
3D building
mesh
create footprint
surface
building
footprint
polygons
mesh footprint
surface
roof-wall
mesh
join and weld
mesh
footprint
mesh
Figure 1. Flowchart of 3D building modeling from ALS point
cloud and polygon footprints
In a first step, we identify the points of the area point cloud that
lie inside the footprint polygons and subsequently associate these
points with the corresponding footprint. This results in a ”small”
point cloud of every building, which we will call the building
point cloud in the following. We now describe the inside of the
footprint polygon with a surface (Figure 2a). We mesh the surface
with the same density as the building point cloud (Figure 2b).
Now we extract only the naked (outer) vertices of the footprint
mesh.
Furthermore we project the building point cloud on the global XY
plane in which also the footprint mesh lies (figure 2c). In this 2D
space, the Delaunay triangulation provides a fast triangulation of
a number of points N. We apply a Delaunay triangulation to the
points, but only use edges that lie inside the footprint polygon
(Figure 2d). Now we glue the upper mesh, consisting of roof
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLII-4/W10, 2018
13th 3D GeoInfo Conference, 1–2 October 2018, Delft, The Netherlands
This contribution has been peer-reviewed.
https://doi.org/10.5194/isprs-archives-XLII-4-W10-179-2018 | © Authors 2018. CC BY 4.0 License.
181
(a) (b)
(c) (d)
(e) (f)
Figure 2. Generation of building mesh: (a) XY plane trimmed
with footprint curve; (b) footprint approximated with mesh; (c)
naked vertices of footprint mesh (green) and building point
cloud projected to XY plane (red); (d) Delaunay mesh of
building (top view); (e) Delaunay triangulation in 3D; (f) refined
and smoothened mesh.
and walls, and the footprint mesh together. Figure 2e shows the
resulting mesh in 3D.
On the building walls we don’t have any points and, thus, end
up with large triangles spanning from the edge of the roof to the
building footprint. This is especially undesirable as the roof-wall
and wall-footprint areas include important shape characteristics
of the building. To combat this, we use a global mesh refinement
that divides every edge of the mesh into kequidistant segments to
generate k2similar smaller triangles (Reuter et al., 2006). Last,
we iteratively apply a form of Laplacian smoothing to the mesh
(Vollmer et al., 1999) to further even out the mesh density on the
walls (Figure 2f). The meshes will now be used to extract the
Shape-DNA of the buildings.
3.2 Shape-DNA
Shape-DNA is a global shape descriptor based on the spectrum
(i.e. the eigenvalues and eigenfunctions) of the Laplace-Beltrami
operator (LBO) of 2D and 3D manifolds. In the continuous case,
the LBO is defined as
f:= div(grad f),(1)
where grad and div are the gradient and divergence on the man-
ifold. The Laplacian eigenvalue problem is given as:
f=λf. (2)
The LBO is intrinsic and as a result, it is invariant to isometric
(metric-preserving) deformations of the manifold (Masci et al.,
2015). For this reason it has been used for the analysis and re-
trieval of non-rigid shapes (such as organic objects), where de-
formations are often near isometric (Lian et al., 2011). However,
non-isometric transformations change the spectrum continuously,
so the spectrum may also be adequate for describing the shape of
buildings and subsequently classifying them.
Different discretisations of the LBO exist, such as the linear finite
element method operator (Reuter et al., 2006). This discretisation
can be used on a triangular surface mesh M(V, E , F )with ver-
tices V={1, ..., N }.
i
j
αij
t2
t1
βij
Figure 3. Example of a polygon mesh
The mesh has to be 2-manifold, meaning each interior edge
(i, j)Eis shared by exactly two triangular faces t1and
t2F, and boundary edges belong to exactly one triangular
face (see Figure 3). In this case the LBO is given as an N×N
matrix ∆ = A-1 B, where the stiffness matrix
A(i, j) =
cotαij +cotβij
2(i, j)edge,
X
kN(i)
A(i, k)i=j,(3)
and the mass matrix
B(i, j) = (|t1|+|t2|
12 (i, j)edge,
PkN(i)|tk|
6i=j,(4)
with |ti|being the area of the triangle ti(Reuter et al., 2009a).
The first nNeigenvalues of the LBO can be computed by
performing the generalised eigendecomposition.
Af=λBf,(5)
where f= (f1, ..., fn)is an N×nmatrix containing as columns
the first ndiscretised eigenfunctions and λ=diag(λ1, ..., λn)
is the diagonal matrix of the corresponding eigenvalues (Masci
et al., 2015). A truncated version of λcan be used as a feature
vector that describes the geometric shape of the underlying mesh.
It is yet unclear as to what number of eigenvalues, i.e. n, should
be used to form the Shape-DNA. Different publications used 11
(Reuter et al., 2009a), 20 (Niethammer et al., 2007) or 10-15
(Lian et al., 2011) eigenvalues for shape analysis and retrieval.
(Arteaga, 2014) defines the LBO directly on a point cloud and
reports about using 50 eigenvalues for accurate shape matching.
(Gao et al., 2014) also discuss this topic and use at most 100
eigenvalues for shape description and (Reuter et al., 2009b) men-
tions that 500 eigenvalues had to be computed for extracting im-
portant information from eigenvalues. In view of signal process-
ing, more eigenvalues contain more information of detail and can
describe the shape more accurately (Gao et al., 2014). However,
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLII-4/W10, 2018
13th 3D GeoInfo Conference, 1–2 October 2018, Delft, The Netherlands
This contribution has been peer-reviewed.
https://doi.org/10.5194/isprs-archives-XLII-4-W10-179-2018 | © Authors 2018. CC BY 4.0 License.
182
more eigenvalues can also carry information about non-isometric
deformations which might make the detection of shape similari-
ties difficult, especially in view of building classification. We will
initially make use of 50 eigenvalues simply because this seems
like a good compromise between the above mentioned numbers
and a feasible amount of eigenvalues to extract in view of data
handling. We will not investigate deeply into the amount of
eigenvalues to be used, however, we do not expect a great in-
fluence when altering this number.
Another problem is that a mesh can only approximate the under-
lying manifold. Simply put, the more dense a mesh, the better
it can approximate the shape. Especially in areas with concavi-
ties or with high surface curvature, most eigenfunctions will need
dense mesh representations (Reuter et al., 2006). However, glob-
ally dense meshes may have a large number of vertices and thus
make the solution of equation 5 difficult or at least very time
consuming. How dense a mesh needs to be for efficient shape
analysis, such as shape classification, is problem dependent and
cannot be said in general. For the building models gained from
ALS point clouds and footprint data, the mesh density poses a
significant problem. As we do not reconstruct the geometry with
parameterised segments, the mesh density is initially limited by
the point density of the ALS point cloud. Additionally, we don’t
have any points on the walls of the building (see section 3.1.2).
As such, we end up with big triangles on the wall segments of the
meshes. This is especially undesirable as the roof-wall and wall-
footprint areas of a building exhibit high surface curvature. Thus,
we refine and smoothen the mesh before extracting Shape-DNA.
Besides being an isometry invariant, Shape-DNA has another
nice property: by normalising it, the spectrum can also be made
scale-invariant. There are different methods for normalising the
spectrum, such as dividing it by its first non-zero eigenvalue or
multiplying it with the surface area of the underlying manifold.
For identifying the roof type of the synthetic building models we
make use of the first method for simplicity reasons. However, be-
cause two similarly shaped buildings with different dimensions
may certainly have different SBSTs scale invariance may not al-
ways be desired.
3.3 Feature extraction
In the following two sections we report on the creation of fea-
ture vectors. We consider all features to be equally important.
Thus, before we use them in a classification, we apply feature
standardisation (Pedregosa et al., 2011), by subtracting the mean
of all feature values and dividing by their corresponding standard
deviation.
3.3.1 Synthetic building models The footprint area and foot-
print perimeter can simply be extracted from the length land
width wof the building models, also regarding extensions. Next,
we extract the gutter height hand the surface area of the mod-
els. Because we constructed the synthetic building models with
piecewise planar segments we could also easily extract the roof
segment count and corresponding roof angle. However, due to
the characteristics of our models, even simple rules would allow
to classify the buildings with these features. Instead we aim to
classify the models only by using Shape-DNA and not using any
explicit information about the roof structure in the classification.
Thus, we extract the first 50 normalised eigenvalues of the meshes
according to the method presented in section 3.2.
3.3.2 Building models from remote sensing and cadastral
data From the Groningen building models we extract the same
features as for the synthetic models. Because, we have the build-
ing footprint as a polygon, the extraction of footprint area and
perimeter is trivial. We approximate the gutter height of the build-
ing as the mean Z coordinate of the roof points. By applying a
region growing segmentation (Grilli et al., 2017) to the building
point cloud we extract the number of roof segments and the aver-
age segment angle weighted by the number of points in a segment
(Figure 4).
Figure 4. Region growing segmentation applied to building
point cloud to extract roof segments
Furthermore, we approximate the buildings surface area as the
sum of all triangle areas of the mesh. Last, we extract Shape-
DNA from the building mesh. The only non-geometric feature
we use for the SBST classification is the year of construction,
stemming from a cadastral dataset.
3.4 Labeling according to Global Earthquake Model
Within the context of the Groningen Earthquakes Structural Up-
grading project, Arup developed a methodology to gain six out
of the 13 GEM attributes for around 20.000 buildings. These are
(Brzev et al., 2013):
(1) The lateral load-resisting system (LLRS) describing the struc-
tural system that provides resistance against horizontal earth-
quake forces through vertical and horizontal components. Ex-
amples are wall based lateral load resisting systems (LWAL) or
hybrid lateral load resisting systems (LH). In the latter case, dif-
ferent types of LLRS are combined in one structure. (2) The
material of the LLRS. Examples may be concrete (CR), wood
(W) or unreinforced masonry (MUR). (3,4) Both the LLRS and
its material are given for the two main directions of the building.
A common example are terraced houses that may have a wall
based lateral load resisting system (LWAL) in the direction par-
allel to the street, but have no lateral load resisting system (LN)
in the direction orthogonal to that. (5) The floor attribute describ-
ing floor material, floor system type, and floor-wall connection.
In this paper we only describe the material of the building floor,
such as wooden floor (FW), concrete floor (FC), masonry floor
(FM) or other floor (FO) material. (6) The material of exterior
walls describing the building enclosure. In this paper we use this
parameter to denote the presence of an outer leaf wall as follows:
outer leaf wall present (EW) or no outer leaf wall present (EWN).
By first expressing the material and LLRS in the main direction,
material and LLRS in the second direction, floor and exterior wall
type, a SBST can be represented such as the following: MUR-
LW AL MUR LN F W EW . This is a common
SBST for a terraced building unit. For the Groningen buildings
these six attributes reflect the most influential parameters for as-
sessing their seismic vulnerability. For several reasons it is not
always possible to identify each of the six attributes for every
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLII-4/W10, 2018
13th 3D GeoInfo Conference, 1–2 October 2018, Delft, The Netherlands
This contribution has been peer-reviewed.
https://doi.org/10.5194/isprs-archives-XLII-4-W10-179-2018 | © Authors 2018. CC BY 4.0 License.
183
building in the labeling process. We consider the distinct com-
binations of the attributes as one class label and therefore only
use buildings for the training set that are fully labeled with all six
attributes. Otherwise there would be classes such as unknown-
LH unknown LH F W E W N. A way to circum-
vent this problem and increase the labeled training set could be to
predict each of the six GEM attributes separately. Although this
is possible in a so-called multi-label classification, we decided to
not use this technique in this paper. The reason for this is that pat-
terns formed by classes represented with the combination of sev-
eral attributes may be easier to detected than those from classes
grouped by single GEM attributes. As an example, a building
with concrete walls can still occur in many different shapes, while
the combination with other GEM attributes describes the building
more precisely. The distinct combination of the attributes results
in more than 30 classes present in our dataset. However, most of
these classes are only represented with around 10 to 100 samples.
The input of the labeling process stems from different sources,
including detailed inspections of building plans and in situ in-
spections. Furthermore, a dataset containing every agricultural
building in the area is included. Stemming from the dataset in-
cluding only agricultural buildings, an educated guess is made
towards the SBST of these buildings. All buildings in this dataset
are labeled with the GEM combination MUR LH MUR-
LH F W EW N . This guess is not always correct and may
therefore also influence the measured classification accuracy.
In a previous classification all buildings that are directly adjacent
to a similar building were identified. These are terraced houses
which are assigned to have no lateral load resisting system (LN)
in the second direction. Because we are interested in expected
classification performance also for many other types, we do not
include an adjacency feature in our experiments as it would im-
mediately identify six out of the eight SBST classes as terraced
houses.
4. EXPERIMENTS
For the following experiments we first use the design application
Rhinoceros 3D and the visual scripting add-on Grasshopper to
construct the synthetic as well as real-world building models as
described in section 3.1. The data for the real-world building
models, namely the cadastral dataset Basisregistratie Adressen
en Gebouwen5and the ALS point cloud Actueel Hoogtebestand
Nederland6, is openly available at the Dutch national geoportal
Publieke Dienstverlening Op de Kaart7. We remesh some of the
models and extract the first 50 eigenvalues of the LBO of the
meshes. Extracting a smaller number of eigenvalues does not
significantly change the outcomes. Additionally, we normalise
the eigenvalues in the first two experiments. For the last three
steps we use the ’ShapeDNA-tria’ software8.
4.1 Influence of meshing on Shape-DNA
With this experiment we aim to answer the question, how den-
sity and quality of a building mesh influence Shape-DNA. We
use the extracted Shape-DNA as feature vectors of the building
models and plot the results using multidimensional scaling (Fig-
ure 5) - a way to visualise high dimensional feature spaces. The
5https://www.kadaster.nl/bag
6http://www.ahn.nl/index.html
7https://www.pdok.nl/
8http://reuter.mit.edu/software/shapedna
red markers represent the building models that were meshed with
0.2m×0.2mgrid spacing. These meshes represent the underly-
ing manifold most accurately. The blue markers represent models
that resemble the meshes gained from remote sensing data. We
can see that the Shape-DNA extracted from these models is far
from the red markers and therefore considered to be imprecise.
By applying mesh refinement and smoothing to these models,
we can increase the precision of the Shape-DNA greatly (cyan
markers). Thus, we conclude that it is possible to extract reason-
ably precise Shape-DNA from smoothened building meshes, that
can also be gained from remote sensing data. However, the cyan
markers still seam to produce a slightly different pattern then the
red and green markers. In this case it also has the effect that the
distance of the cyan markers in this embedding between gabled
and flat roofed buildings is slightly smaller. If this poses an issue
for classification will be answered in the next experiment.
Figure 5. MDS plot of Shape-DNA of synthetic building
models. The legend shows the grid spacing of the meshing and
different roof types of the models. It is clearly visible that
remeshing and smoothing of the models without wall points
increases the accuracy of the extracted Shape-DNA.
4.2 Roof type prediction of synthetic building models
With this experiment we aim to answer the question, how Shape-
DNA performs at predicting the roof type of a building. Again,
we use the synthetic models that resemble the meshes gained
from remote sensing data. For the classification, we use a SVM
with Gaussian radial basis function kernel, as implemented in the
Python machine learning library Scikit-learn (Pedregosa et al.,
2011). Table 1 shows an overview of all the carried out classi-
fications. The average classification accuracy is measured using
10 fold cross-validation.
!
!
!
!
!
!
!
!
!
Samples(per(class(
Extension(in(!"(
Accuracy(in(%(
$100!
$0!
93.2!
$100!
1!
90.1!
$100!
4!
79.7!
$100!
9!
75.0!
$100!
{$0, 1, 4, 9}!
66.3!
$200!
{$0, 1, 4, 9}!
74.3!
$300!
{$0, 1, 4, 9}!
77.6!
$400!
{$0, 1, 4, 9}!
79.6!
!
!
!
(93.2+77.9+79.7+75)/4=81.45!
Table 1. Results of roof type classification
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLII-4/W10, 2018
13th 3D GeoInfo Conference, 1–2 October 2018, Delft, The Netherlands
This contribution has been peer-reviewed.
https://doi.org/10.5194/isprs-archives-XLII-4-W10-179-2018 | © Authors 2018. CC BY 4.0 License.
184
For buildings without an extension we can predict the roof type
with 93.2% average classification accuracy. With an increasing
size of the extension the predictive accuracy decreases. While
an extension of 1m×1m×1m, e.g. representing a chimney,
does not affect the performance significantly, bigger extensions
do. For models with an extension of 3m×3m×3m, e.g. rep-
resenting a small garage, the average classification accuracy goes
down to 75.0%. This is mostly due to buildings with smaller
dimensions, e.g. 4m×5m×3m, where a big extension signifi-
cantly changes the building shape. Thus, the Shape-DNA features
are less efficient in capturing local changes on the roof. A local
shape descriptor only describing the (main) roof shape would not
be affected by an extension. For four more experiments we ran-
domly draw 100, 200, 300 and 400 buildings per roof type with
all possible extension types. We can see that the classification
significantly benefits from more training data. With 400 build-
ings per class we achieve an average accuracy of 79.6%. This
proves that the SVM classifier can become partially invariant to
the building extensions. Such a behaviour is desirable when clas-
sifying buildings according to their shape. Even for buildings
with different dimensions and extensions the roof type can still
be predicted.
In a different experiment we also classify the models according
to the presence and size of their extension. Here we reach an av-
erage accuracy of 92.3% over all four classes. This proves again
that Shape-DNA is a relevant shape descriptor for building shape
classification. Additionally, we also conduct some of the previ-
ous experiments with meshes with a 0.2m×0.2mgrid spacing,
where we often gain around 3% predictive accuracy. In general,
we conclude that it is possible to predicted the roof type and other
shape features, such as the presence of an extension, by using
Shape-DNA on 3D building models that can also be gained from
remote sensing data.
4.3 SBST prediction for Groningen building models
With this experiment we aim to answer the question, how the
features extracted in section 3.3.2 perform at predicting GEM at-
tributes of real-world buildings. From the dataset described in
section 3.4 we only use the eight most common classes to get
a reliable measure of the classification accuracy. To provide an
equal number of training samples for each class we randomly un-
dersample the classes to 100 buildings each. In this experiment,
we make use of a RF to classify the buildings, as it performs
better compared to a SVM. Again, we measure average classi-
fication accuracy with a 10 fold cross-validation using different
feature combinations. The RF also allows to inspect the feature
importance score in the SBST classification. According to this,
the year of construction and footprint area are the most important
features. We will further discuss this when inspecting a confu-
sion matrix of the classification. Table 2 shows an overview of
the average classification accuracy of different feature sets.
2D#
3D#
Semantic#
Accuracy#in#%#
fparea#
peri-
meter#
gutter#
height#
rcount#
rangle#
sarea#
Shape#
DNA#
yoc#
!
!
!
!
!
!
!
!
60.1!
!
!
!
!
!
!
!
!
75.5!
!
!
!
!
!
!
!
!
78.3!
!
!
!
!
!
!
!
!
79.6!
!
!
!
!
!
!
!
!
75.3!
!
Table 2. Results of SBST classification
We achieve 75.5% average classification accuracy only using
footprint area (fparea), perimeter and year of construction (yoc)
features in the classification. These three features are all explic-
itly included in the cadastral dataset. Adding 3D features does not
significantly improve the classification accuracy. Adding Shape-
DNA even results in a slightly worse performance. This may be
due to the fact that we only have 100 training samples per class,
but potentially a variety of small shape variations inside a class.
It would be interesting to see if the performance of Shape-DNA
improves with more training data, similar to the behaviour we so
on the synthetic models.
We inspect a confusion matrix of the best performing feature set
to gain further inside (Figure 6). We can see that the masonry
terraced houses (M UR LW AL MUR LN XE W )
achieve the lowest accuracy. The three classes with this type
only differ in their floor material, which may not form patterns
that are visible in the shape or age of the buildings. We still
achieve good accuracies, most likely due to patterns formed by
common building practice. Terraced houses with concrete walls
(CR LW AL CR LN XX) achieve the highest
classification accuracy. These buildings can mainly be identi-
fied because they were constructed more recently compared to
the masonry terraced units. Agricultural (MURLH MUR-
LH F W EW N ) and single houses (M U R LW AL-
MU R LWAL F W EW N ) can often be identified by
their larger footprint areas and thus also achieve a high classifica-
tion accuracy.
Figure 6. Confusion matrix of GEM classification.
5. CONCLUSIONS AND FUTURE WORK
In this paper we apply a new method to describe and classify
buildings by using Shape-DNA. On synthetic building data, we
show that it is possible to predict the roof type of buildings and
the presence of an extension only by using Shape-DNA. For the
real-world data from the Groningen region we predict SBSTs de-
fined by the GEM building taxonomy. We achieve high accura-
cies, only using simple features, such as the footprint area and
year of construction from a 2D cadastral dataset. We thereby val-
idate previous research in SBST prediction for a sample of the
Groningen building stock. Although we only predict attributes
for less than 1% of the building stock in this paper, our method
can in theory be applied for the entire region and possibly also
for different regions. However, we can only predict SBSTs that
are present in our training set. In the future we want to investi-
gate into predicting all the classes of this dataset and improve the
classification accuracy for classes that do not achieve a satisfying
performance yet. We expect to achieve both by using the remain-
ing training data and potentially adding different feature types,
such as visual or geographic features to the classification. We
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLII-4/W10, 2018
13th 3D GeoInfo Conference, 1–2 October 2018, Delft, The Netherlands
This contribution has been peer-reviewed.
https://doi.org/10.5194/isprs-archives-XLII-4-W10-179-2018 | © Authors 2018. CC BY 4.0 License.
185
also tried to incorporate the shape features gained by Shape-DNA
in the presented SBST classification. However, Shape-DNA was
not able to improve the classification performance in our exper-
iments. Local shape features, such as the footprint area or the
number of roof segments led to better results, provided these fea-
tures can be extracted. Shape-DNA may require a larger number
of training samples, which is often problematic and particularly
so in SBST classification. In situations where a large amount
of training data is available and the aim is to classify buildings
into uniform geometric groups, Shape-DNA can still be useful.
Furthermore, Shape-DNA could be useful in shape retrieval prob-
lems, by using the spectrum of real-world buildings to find similar
building models, e.g. in a 3D model database.
In general our paper shows that it is possible to predict building
attributes according to the GEM taxonomy by using simple clas-
sification models and, thus, provide important input for rapid and
precise large scale seismic assessment.
ACKNOWLEDGEMENTS
This paper describes an activity that has been partly developed
in the Groningen Earthquakes Structural Upgrading Arup project
for the Nederlandse Aardolie Maatchapij BV. The authors would
like to thank all the parties and colleagues involved in the project.
REFERENCES
Arteaga, R. J., 2014. Laplace-beltrami spectra as ‘shape-dna’ of
surfaces using the closest point method. Master’s thesis, Simon
Fraser University.
Borfecchia, F., Pollino, M., Cecco, L. D., Lugari, R., Martini,
R., Porta, L. L., Ristoratore, E. and Pascale, C., 2010. Active
and passive remote sensing for supporting the evaluation of the
urban seismic vulnerability. Italian Journal of Remote Sensing
pp. 129–141.
Borzi, B., Dell’Acqua, F., Faravelli, M., Gamba, P., Lisini, G.,
Onida, M. and Polli, D., 2011. Vulnerability study on a large
industrial area using satellite remotely sensed images. Bulletin of
Earthquake Engineering 9(2), pp. 675–690.
Brzev, S., Scawthorn, C., Charleson, A. W., Allen, L., Greene,
M., Jaiswal, K. and Silva, V., 2013. Gem building taxonomy
(version 2.0). Technical report, GEM Foundation.
Christodoulou, A., Kokkos, A. and Palmieri, M., 2017. Auto-
mated building stock data mining and classification using open
source data. In: IASS 2017, At Hamburg, Germany.
Gao, Z., Yu, Z. and Pang, X., 2014. A compact shape descrip-
tor for triangular surface meshes. Computer-Aided Design 53,
pp. 62–69.
Geiß, C., 2015. Seismic vulnerability assessment of built envi-
ronments with remote sensing. http://dx.doi.org/10.18452/17104,
Humboldt-Universit¨
at zu Berlin.
Geiß, C., Aravena Pelizari, P., Marconcini, M., Sengara, W., Ed-
wards, M., Lakes, T. and Taubenboeck, H., 2015. Estimation
of seismic building structural types using multi-sensor remote
sensing and machine learning techniques. ISPRS Journal of Pho-
togrammetry and Remote Sensing 104, pp. 175–188.
Grilli, E., Menna, F. and Remondino, F., 2017. A REVIEW OF
POINT CLOUDS SEGMENTATION AND CLASSIFICATION
ALGORITHMS. Vol. XLII-2/W3.
Haala, N. and Kada, M., 2010. An update on automatic 3d build-
ing reconstruction. ISPRS Journal of Photogrammetry and Re-
mote Sensing 65(6), pp. 570–580.
Lian, Z., Godil, A., Bustos, B., Daoudi, M., Hermans, J., Kawa-
mura, S., Kurita, Y., Lavou´
e, G., Nguyen, H. V., Ohbuchi, R.,
Ohkita, Y., Ohishi, Y., Porikli, F., Reuter, M., Sipiran, I., Smeets,
D., Suetens, P., Tabia, H. and Vandermeulen, D., 2011. SHREC
’11 Track: Shape Retrieval on Non-rigid 3D Watertight Meshes.
In: H. Laga, T. Schreck, A. Ferreira, A. Godil, I. Pratikakis and
R. Veltkamp (eds), Eurographics Workshop on 3D Object Re-
trieval, The Eurographics Association.
Lugari, A., 2014. Active and passive remote sensing techniques
and artificial neural networks in support of buildings seismic vul-
nerability assessment. PhD thesis, TOR VERGATA UNIVER-
SITY, ROME, ITALY.
Masci, J., Boscaini, D., Bronstein, M. M. and Vandergheynst,
P., 2015. Geodesic convolutional neural networks on riemannian
manifolds. In: 2015 IEEE International Conference on Computer
Vision Workshop (ICCVW), pp. 832–840.
Muntendam-Bos, A. G., Roest, J. P. A. and De Waal, H., 2015. A
guideline for assessing seismic risk induced by gas extraction in
the Netherlands. Vol. 34, Society of Exploration Geophysicists.
Nederlandse Aardolie Maatschappij BV, 2015. Hazard and risk
assessment for induced seismicity groningen. interim update 7
november 2015. Technical report.
Niethammer, M., Reuter, M., Wolter, F.-E., Bouix, S., Peinecke,
N., Koo, M.-S. and Shenton, M. E., 2007. Global Medical Shape
Analysis Using the Laplace-Beltrami Spectrum. Springer Berlin
Heidelberg, Berlin, Heidelberg, pp. 850–857.
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion,
B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg,
V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Per-
rot, M. and Duchesnay, ´
E., 2011. Scikit-learn: Machine Learning
in Python. Journal of Machine Learning Research.
Pittore, M. and Wieland, M., 2013. Toward a rapid probabilistic
seismic vulnerability assessment using satellite and ground-based
remote sensing. Natural Hazards 68(1), pp. 115–145.
Reuter, M., Biasotti, S., Giorgi, D., Patan`
e, G. and Spagnuolo,
M., 2009a. Discrete laplace–beltrami operators for shape analysis
and segmentation. Computers and Graphics 33(3), pp. 381 –
390. IEEE International Conference on Shape Modelling and
Applications 2009.
Reuter, M., Wolter, F.-E. and Peinecke, N., 2006. Laplace-
beltrami spectra as ’shape-dna’ of surfaces and solids. Comput.
Aided Des. 38(4), pp. 342–366.
Reuter, M., Wolter, F.-E., Shenton, M. and Niethammer, M.,
2009b. Laplace–beltrami eigenvalues and topological features of
eigenfunctions for statistical shape analysis. Computer-Aided De-
sign 41(10), pp. 739 – 755. Selected Papers from the 2007 New
Advances in Shape Analysis and Geometric Modeling Workshop.
Sarabandi, P., 2007. Development of algorithms for building in-
ventory compilation through remote sensing and statistical infer-
encing. PhD thesis, Stanford University.
Vollmer, J., Mencl, R. and Muller, H., 1999. Improved laplacian
smoothing of noisy surface meshes. Computer Graphics Forum
18(3), pp. 131–138.
Wieland, M., Pittore, M., Parolai, S., Zschau, J., Moldobekov,
B. and Begaliev, U., 2012. Estimating building inventory for
rapid seismic vulnerability assessment: Towards an integrated ap-
proach based on multi-source imaging. Soil Dynamics and Earth-
quake Engineering 36(Supplement C), pp. 70 – 83.
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLII-4/W10, 2018
13th 3D GeoInfo Conference, 1–2 October 2018, Delft, The Netherlands
This contribution has been peer-reviewed.
https://doi.org/10.5194/isprs-archives-XLII-4-W10-179-2018 | © Authors 2018. CC BY 4.0 License.
186
Article
The seismic building structural type (SBST) represents critical information for accurately assessing seismic risk. Such risk is typically determined as a function of seismic hazards, assets such as buildings and people exposed to hazards, and the vulnerability of those exposed elements (especially buildings because SBSTs dictate the main load-bearing structures of buildings and their seismic vulnerability). China has vast rural areas that generally differ in building upgrade status. SBST distribution information for these areas is often outdated, missing, or nonexistent, thus hindering local governments’ planning for earthquake preparedness and mitigation. This study presents a low-cost method for estimating SBSTs in large rural areas. According to building features (e.g., footprint, roof type) extracted from remote sensing data and building-related local knowledge (BRLK) expressed as building group pattern-SBST correlations, a series of SBST recognition rules is established using the decision tree model and statistical methods. The results from experimental verification show the following. Using only decision tree-based recognition, the estimation of SBSTs reaches an overall accuracy of approximately 70%. After combining decision tree-based recognition with BRLK, the overall estimation accuracy exceeds 90%. The proposed method establishes reliable recognition rules for estimating SBSTs in large rural areas and has a strong potential for popularization and application. In the future, the proposed method can be further investigated according to the BRLK in various areas in terms of applicability and adjustment schemes.
Conference Paper
Full-text available
The rising attention on the sustainability and resilience of the built environment requires an enhanced understanding of the existing building stock. The number and diversity of buildings that are necessary to analyse in order to assess the performance of a region's building stock justifies the development of automated building information harvesting processes. This paper will present an overview of the automated data-mining processes that have been used to analyse and classify the building stock within the context of the Groningen Earthquakes Structural Upgrading Arup project. Building information has been extracted from available public data sources, namely: building outline data from the building outline database of Kadaster [1], and height map data from the AHN database (Actueel Hoogtebestand Nederland – Netherlands Height Map) of PDOK (Publieke Dienstverlening Op de Kaart – Dutch National Spatial Data Infrastructure) [2]. Arup has developed algorithmic tools that process this open source data in order to automatically calculate building-by-building properties and therefore classify buildings into groups with relative geometric uniformity. Within these building groups, the study of predominant structural characteristic combination typologies is facilitated along with their expected count within the study area. This process reduces the uncertainty on the overall resilience assessment leading to better informed prioritisation & decision making.
Article
Full-text available
In the Netherlands, seismicity is induced by the reactivation of faults because of the extraction of gas. The Dutch mining law requires a seismic-risk assessment as part of the license application process. For this purpose, a risk-assessment guideline has been developed over the past decade. The guideline contains three assessment levels. At the first level, a screening occurs to assess the potential of inducing seismicity. On the basis of three key parameters and an analysis of the maximum potential magnitude, each field can be classified for induced-seismicity risk prior to the onset of production. For fields with a low seismicity potential, the existing national monitoring network suffices. At the second level, for fields with medium and high seismicity potential, a qualitative assessment of hazard and risk is required based on a risk-matrix approach. The large Groningen gas field is considered a field with high seismic risk requiring a level 3 assessment. For such fields, a probabilistic seismic-risk assessment and risk-management plan are required. The risk assessment requires special attention because most of the seismic risk is associated with low-probability events that can induce large ground accelerations.
Article
Full-text available
A patent application based on the key ideas of this paper was submitted in June 2005 by the authors of this paper as inventors with F.-E. Wolter holding (50 %), N. Peinecke (25%) and M. Reuter (25%) of the respective intellectual property rights of this patent which was approved in 2009 as US Patent App. 11/916,104, 2009 cf. https://patents.google.com/patent/US20090169050A1/en This paper introduces a method to extract ‘Shape-DNA’, a numerical fingerprint or signature, of any 2d or 3d manifold (surface or solid) by taking the eigenvalues (i.e. the spectrum) of its Laplace–Beltrami operator. Employing the Laplace–Beltrami spectra (not the spectra of the mesh Laplacian) as fingerprints of surfaces and solids is a novel approach. Since the spectrum is an isometry invariant, it is independent of the object’s representation including parametrization and spatial position. Additionally, the eigenvalues can be normalized so that uniform scaling factors for the geometric objects can be obtained easily. Therefore, checking if two objects are isometric needs no prior alignment (registration/localization) of the objects but only a comparison of their spectra. In this paper, we describe the computation of the spectra and their comparison for objects represented by NURBS or other parametrized surfaces (possibly glued to each other), polygonal meshes as well as solid polyhedra. Exploiting the isometry invariance of the Laplace–Beltrami operator we succeed in computing eigenvalues for smoothly bounded objects without discretization errors caused by approximation of the boundary. Furthermore, we present two non-isometric but iso spectral solids that cannot be distinguished by the spectra of their bodies and present evidence that the spectra of their boundary shells can tell them apart. Moreover, we show the rapid convergence of the heat trace series and demonstrate that it is computationally feasible to extract geometrical data such as the volume, the boundary length and even the Euler characteristic from the numerically calculated eigenvalues. This fact not only confirms the accuracy of our computed eigenvalues, but also underlines the geometrical importance of the spectrum. With the help of this Shape-DNA, it is possible to support copyright protection, database retrieval and quality assessment of digital data representing surfaces and solids. Keywords: Laplace–Beltrami operator; Shape invariants; Fingerprints; Shape matching; Database retrieval; Copyright protection; NURBS; Parameterized surfaces and solids; Polygonal meshes
Article
Full-text available
Detailed information about seismic building structural types (SBSTs) is crucial for accurate earthquake vulnerability and risk modeling as it reflects the main load-bearing structures of buildings and, thus, the behavior under seismic load. However, for numerous urban areas in earthquake prone regions this information is mostly outdated, unavailable, or simply not existent. To this purpose, we present an effective approach to estimate SBSTs by combining scarce in situ observations, multi-sensor remote sensing data and machine learning techniques. In particular, an approach is introduced, which deploys a sequential procedure comprising five main steps, namely calculation of features from remote sensing data, feature selection, outlier detection, generation of synthetic samples, and supervised classification under consideration of both Support Vector Machines and Random Forests. Experimental results obtained for a representative study area, including large parts of the city of Padang (Indonesia), assess the capabilities of the presented approach and confirm its great potential for a reliable area-wide estimation of SBSTs and an effective earthquake loss modeling based on remote sensing, which should be further explored in future research activities.
Article
Full-text available
Natural hazards such as earthquakes threaten millions of people all around the world. In a few decades, most of these people will live in fast-growing, inter-connected urban environments. Assessing risk will, therefore, be an increasingly difficult task that will require new, multidisciplinary approaches to be tackled properly. We propose a novel approach based on different imaging technologies and a Bayesian information integration scheme to characterize exposure and vulnerability models, which are among the key components of seismic risk assessment.
Thesis
Globale Urbanisierungsprozesse und eine Zunahme der räumlichen Konzentration von exponierten Elementen wie Menschen, Gebäude, Infrastruktur und ökonomische Werte induzieren ein ungekanntes Risiko in erdbebengefährdeten Regionen. Wenn keine Abschwächung des Risikos erfolgt werden dramatische Folgen in der Zukunft erwartet. Diese umfassen eine beispiellose Anzahl an Todesopfer, enorme ökonomische und ökologische Verluste und Ausfälle bezüglich kritischer Infrastruktur und Versorgung etc. Um derartige Gefährdungen abzuschwächen sind detaillierte Informationen über seismisches Risiko notwendig. Die seismische Verwundbarkeit von Siedlungsarealen ist dabei als zentrale, konstituierende Komponente von seismischem Risiko zu berücksichtigen. In diesem Zusammenhang ist es von besonderem Interesse das Verhalten von Gebäudeinventaren unter einem bestimmten Erdbebeneinfluss abschätzen zu können. Das Hauptziel der Arbeit war es maßgeschneiderte Methoden zu entwickeln, die eine Bewertung der seismischen Vulnerabilität von Siedlungsräumen, basierend auf Fernerkundungsdaten, durchführbar machen. Es wurden Methoden aus dem Bereich des maschinellen Lernens adaptiert, um Verwundbarkeitsstufen von Gebäuden und homogenen Siedlungsstrukturen zu bestimmen. Hierfür wurden Merkmale aus Fernerkundungsdaten abgeleitet und mit in situ Informationen verknüpft. Wir verwenden verschiedene Ensembles von Fernerkundungssensoren, um die urbane Morphologie umfassend zu charakterisieren. Empirische Ergebnisse, die für die erdbebengefährdeten Städte Padang (Indonesien) und Istanbul (Türkei) generiert werden konnten, bestätigen die Durchführbarkeit der entwickelten Verfahren. Zukünftige Arbeiten können daran anknüpfen und beispielsweise empirische Erkenntnisse in weiteren Fallstudien anzweifeln, eine Verbesserung der Methodik vornehmen, Konzepte und Ansätze auf andere Sensorsysteme oder Datenquellen übertragen oder Daten und Methoden im Rahmen von holistischen Risikobewertungsstrategien anwenden.
Article
Three-dimensional shape-based descriptors have been widely used in object recognition and database retrieval. In the current work, we present a novel method called compact Shape-DNA (cShape-DNA) to describe the shape of a triangular surface mesh. While the original Shape-DNA technique provides an effective and isometric-invariant descriptor for surface shapes, the number of eigenvalues used is typically large. To further reduce the space and time consumptions, especially for large-scale database applications, it is of great interest to find a more compact way to describe an arbitrary surface shape. In the present approach, the standard Shape-DNA is first computed from the given mesh and then processed by surface area-based normalization and line subtraction. The proposed cShape-DNA descriptor is composed of some low frequencies of the discrete Fourier transform of the processed Shape-DNA. Several experiments are shown to illustrate the effectiveness and efficiency of the cShape-DNA method on 3D shape analysis, particularly on shape comparison and classification.
Article
We propose an integrated approach to estimating building inventory for seismic vulnerability assessment, which can be applied to different urban environments and be efficiently scaled depending on the desired level of detail. The approach employs a novel multi-source method for evaluating structural vulnerability-related building features based on satellite remote sensing and ground-based omnidirectional imaging. It aims to provide a comparatively cost- and time-efficient way of inventory data capturing over large areas. The latest image processing algorithms and computer vision techniques are used on multiple imaging sources within the framework of an integrated sampling scheme, where each imaging source and technique is used to infer specific, scale-dependent information. Globally available low-cost data sources are preferred and the tools are being developed on an open-source basis to allow for a high degree of transferability and usability. An easily deployable omnidirectional camera-system is introduced for ground-based data-capturing. After a general description of the approach and the developed tools and techniques, preliminary results from a first application to our study area, Bishkek, Kyrgyzstan, are presented.