Professional Documents
Culture Documents
1 School of Geomatics and Urban Spatial Informatics, Beijing University of Civil Engineering and
Architecture, Beijing 100044, China; dongyouqiang@bucea.edu.cn (Y.D.); houmiaole@bucea.edu.cn (M.H.);
liyihao970721@163.com (Y.L.); ji_yuhang@163.com (Y.J.)
2 Chinese Academy of Surveying and Mapping, Beijing 100036, China
Abstract: The Ming and Qing Dynasty type of official-style architecture roof can provide plenty of
prior knowledge relating to the structure and size of these works of architecture, and plays an im-
portant role in the fields of 3D modeling, semantic recognition and culture inheriting. In this paper,
we take the 3D point cloud as the data source, and an automatic classification method for the roof
type of Ming and Qing Dynasty official-style architecture based on the hierarchical semantic net-
work is illustrated. To classify the roofs into the correct categories, the characteristics of different
roof types are analyzed and features including SoRs, DfFtR, DoPP and NoREs are first selected;
subsequently, the corresponding feature extraction methods are proposed; thirdly, aiming at the
structure of the ridges, a matching graph relying on the attributed relational graph of the ridges is
given; based on the former work, a hierarchical semantic network is proposed and the thresholds
are determined with the help of the construction rules of the Ming and Qing Dynasty official-style
Citation: Dong, Y.; Hou, M.; Xu, B.; architecture. In order to fully verify the efficiency of our proposed method, various types of Ming
Li, Y.; Ji, Y. Ming and Qing Dynasty and Qing Dynasty official-style architecture roof are identified, and the experimental results show
Official-Style Architecture Roof that all structures are classified correctly.
Types Classification Based on the 3D
Point Cloud. ISPRS Int. J. Geo-Inf. Keywords: 3D point cloud; roof classification; construction rules; features selection; hierarchical
2021, 10, 650. https://doi.org/10.3390/ semantic network
ijgi10100650
Roof types have been widely used in 3D building and parametric modeling and have
led to many achievements. Kada and McKinley [10] firstly segmented the point cloud and
determined that the roof type relies on the number of segments; then, the roof plane pa-
rameters were estimated. Mass and Vosselman [11] selected the heights of the point cloud
as weight functions in moment equations to compute the information on the roof type and
shape parameters. On this basis, the gable roof was reconstructed in [11]. Much of the
literature indicates that roof types play an important role in 3D reconstruction [12–15].
Similarly, the roof types (form) of the Ming and Qing Dynasty official-style architec-
ture can also improve the efficiency of parametric modeling. Moreover, in the Ming and
Qing Dynasty, the hierarchy was strictly divided and the construction of the Ming and
the Qing Dynasty official-style architecture followed certain rules, which can be searched
for in YingzaoFashi (Building Standards) [16] of the Song Dynasty or Gongchengzuofa
zeli (Structural Regulations) [17] published by Qing. The roof types (form) of the Ming
and Qing Dynasty official-style architecture can provide more information, including the
structure, topological relation and the size of the components. Based on the information
from the roof types, Yuan Shen [18] calculated the parameters of the traditional Chinese
curvilinear roof based on the roof types (form). Other 3D reconstruction methods for an-
cient Chinese architecture using roof types can be seen in [19–21].
Hence, understanding how to distinguish the roof types of the Ming and Qing Dyn-
asty official-style architecture from the point cloud is an important task, and has a great
significance in the field of 3D modeling.
Bittner et al. [31] propose a Multi-Task conditional generative adversarial network (cGAN)
for simultaneous space borne DSM refinement and roof-type classification.
The methods based on deep learning for roof classification shows promising results;
however, there are some limitations. On one hand, although some roof type training da-
tasets have been published [32], these datasets mainly consist of images and there is rarely
a point cloud dataset focusing on the Ming and Qing Dynasty official-style architecture
roof. On the other hand, the direct method of applying deep learning on a point cloud is
to convert the data into volume representation. Comparing with the 2D image, the amount
of 3D volume data will become very large, very fast (although a graphics processing unit
(GPU) has been developed). To speed up processing, it is necessary to compromise and
adopt a lower resolution (some methods use 64 × 64 × 64). This strategy will cause the loss
of details and bring the cost of quantization errors. In fact, some experimental results show
that the classical methods may obtain better performance than the deep learning methods
in the field of architectural heritage point cloud classification [33].
are a few examples. The main difference of these methods is what kind of features are
selected and extracted to construct the topological graph. For example, Cheng et al. [47]
select various curvature information to construct the topological graph; Hao et al. [48]
defines the connection types between different types of planes and analyzes the structure
of common objects; Berner et al. [49] combines the outlines and shape to construct the
topological graph. For the Chinese ancient architecture roof, the structure of the roof is the
important feature. However, only relying the structure of the different Ming and Qing
Dynasty official-style architectures is impossible. This is mainly because the different roof
types may have the same structure relationship. For example, the structure of the over-
hanging gable roof and the flush gable roof are the same.
The object recognition method based on machine learning extracts and learns the fea-
tures of samples, and uses the classifier to complete the classification and recognition of
objects. Generally speaking, machine learning classifiers can be divided into two catego-
ries: traditional machine learning and deep learning. In the traditional machine learning
method, the commonly used classifiers are: support vector machine [50], random forest
[51], adaboost [52], jointboost [53], naive Bayes classifier [54] and maximum expectation
algorithm [55]. Nowadays, the research of point cloud classification mainly focuses on the
recognition of the different ground objects from complex scenes. Much of the literature
shows that the most important impactors affecting recognition are the selected features.
Hence, the extraction of the features presented in the Ming and Qing official buildings
with different styles is very important in the process of the Ming and Qing official building
style classification based on the point cloud.
Thanks to the performance improvement of computer hardware, some architectural
heritage classification methods based on deep learning are proposed. One of the earliest
and most famous deep learning architectures working directly on the point cloud is point
net [56]. It is an end-to-end deep neural network, which can learn the features of classifi-
cation, part segmentation and semantic segmentation by itself. Because the point net does
not capture the local geometry, a development, namely PointNet++ [57] is proposed. Be-
sides the PointNet, some researchers [58,59] applied convolution to gain understanding
of point-based learning. In 2019, Wang et al. propose a method named DGCNN [60]. In-
stead of employing individual points, this method exploits local geometric structures by
constructing a local neighborhood graph and applying convolution-like operations on the
edges connecting neighboring pairs of points. Although the architectural heritage classi-
fication methods based on deep learning have obtained a good performance, no matter
what the ML or DL, the training datasets are very important. Nowadays, some datasets
such as ModelNet 40 [61], KITTI [62], Sydney Urban Objects dataset [63], Semantic3D [64],
S3DIS [65] and ArCH [66] have been published, most of the current datasets collect data
from urban environments, and there are still no published datasets focusing on immova-
ble cultural assets with an adequate level of detail. This brings difficulties for Ming and
Qing Dynasty official-style architecture roof types recognition.
A hierarchical semantic network for the Ming and Qing official-style architecture roof
classification is proposed. In this framework, adaptive thresholds are estimated
based on the construction rule of Qing Dynasty architecture, and the reliable thresh-
olds are given in this paper.
Research in this paper is organized as follows. Section 2 analyzes and selects the iden-
tified features for the classification of the Ming and Qing Dynasty official-style architec-
ture roof. Section 3 gives the corresponding features extraction methods. To recognize the
structure of the ridges, graph matching relying on the attributed relational graph of the
ridges is proposed in Section 4. In Section 5, a hierarchical semantic network for the clas-
sification approach of the Ming and Qing Dynasty official architecture roof is proposed.
Section 6 shows the experimental results and analysis. Finally, a conclusion is conducted
in Section 7.
2. Feature Selection
2.1. A Brief Introduction of the Roof Types
The roof of ancient Chinese architecture plays a particularly important role in the
building facade, which makes the building produce a unique and strong visual effect and
artistic appeal. For the Ming and Qing Dynasty official-style architecture, the roof types
include basic, special and derived roof types. The basic roof types can be classified into
five categories: hip roof, gable-and-hip roof, pyramidal roof, overhanging gable roof, and
flush gable roof as is shown in Table 1. The gable-and-hip roof, gable roof and overhang-
ing gable roof have a corresponding round ridge roof; and the hip roof, gable-and-hip roof
and pyramidal roof can be further divided into double-eave and single-eave as is shown
in Figure 1. Different roof types reflect the different architectural hierarchies and the social
status of the holder. For example, the hip roof can only be used for the royal buildings or
Confucius halls. From high to low rank, the order of different roof types’ grades is: double-
eave hip roof, double-eave gable-and-hip roof, double-eave pyramidal roof, hip roof, ga-
ble-and-hip roof, pyramidal roof, overhanging gable roof, round ridge roof, and flush ga-
ble roof.
Table 1. The roof types of the Ming and Qing official-style architecture.
Gable and hip roofs, with two curving sides, are second in
gable and hip
importance to hip roofs. They are nine ridges including a
roof
main ridges, four vertical ridges and four diagonal ridges.
Flush gable roofs have a main ridge and raise sloping ridges
flush gable roof on the gable walls. It is a very simple style with two slopes
facing front and back.
Besides these basic roof types, there are some special and derived roof types for the
Ming and Qing Dynasty official-style architecture. The common ones are: fan-shaped ga-
ble and hip roof with a round ridge, hip and flat roof, intersecting gable and hip roofs
(Figure 2). These roofs are derived from different combinations of basic roofs. For exam-
ple, the intersecting gable and hip roof is formed by the intersection of two gable-and-hip
roofs. The representative intersecting gable and hip roofs building is the corner building
of the Palace Museum in Beijing. Through various combinations of roofs, the shape and
contour of the building become more abundant.
The roof types of the Ming and Qing Dynasty official-style architecture are rich and
varied. It is difficult to list all the roof types. Compared with the special roof types and the
derived roof types, the basic roof types are more common. Hence, in this paper, we mainly
focus on the classification of the basic roof types.
2.2. Feature Analysis for the Ming and Qing Dynasty Official-Style Architecture Roof
Classification
According to the description in Table 1, the roof of the Ming and Qing Dynasty offi-
cial-style architecture is mainly composed of roofing and ridges. Among the different
types of roofs, the number, topological structure of the ridges and the shape, number, and
topological relationship of irregular surfaces are different. All these differences can be re-
garded as the identified features. Although the features from the ridges and roof surfaces
are coupled with each other, the features from the ridges can provide more details and are
more robust in the recognition of the Ming and Qing Dynasty official-style architecture
compared with the features from roof surface. For example, both the hip roof and the
pyramidal roof have four slopes. The difference between these two types of roofs is
whether there is a main ridge. Similarly, the gable-and-hip roof with a round ridge does
not have a main ridge, while the gable-and-hip roof has a main ridge.
However, only relying on the ridges to classify the roof of the Ming and Qing official-
style architecture into correct categories is impossible. Several of the limitations are listed
as follows:
ISPRS Int. J. Geo-Inf. 2021, 10, 650 7 of 24
The structural relationship of the ridge vector graph from the flush gable roof and
overhanging gable roof are almost the same.
The ridges cannot provide the single-eave or multiple-eave information which is
used to distinguish the single-eave or multiple-eave hip roof, pyramidal roof and
gable and hip roof.
To overcome these limitations, other features from the Ming and Ming and Qing
Dynasty official-style architectures should be considered. Based on these analyses, the se-
lected features for the roof classification are listed as follows.
1. The structure of the ridges—SoRs
The ridges from different roof types of the Ming and Qing Dynasty official-style ar-
chitecture have different topological relationships. The structure of the ridges of different
roof types can be seen in Figure 3. Based on SoRs, most roofs can be classified into the
correct categories.
2. The distance from the outline of the facades to outline of the roof—DfFtR
Figure 4a,b shows a flush gable roof building and an overhanging gable roof build-
ing, respectively. For the flush gable roof building, the outline of the facades is basically
consistent with the outline of roof on the XOY plane as is shown in Figure 4c, while the
outline of the facades is far away from the roof contour on both sides of the overhanging
gable roof building as is shown in Figure 4d. From the perspective of the construction
system for the Ming and Qing Dynasty official-style architecture, the distance from the
outline of the facades to outline of the roof (DfFtR) of the overhanging gable roof building
is also called cantilever length (悬挑). Obviously, DfFtR can be regarded as a salient fea-
ture for distinguishing the flush gable roof and overhanging gable roof.
(a) (b)
(c) (d)
Figure 4. (a) Illustration of a flush gable roof building; (b) illusion of an overhanging gable roof
building; (c) the roof and facades of a flush gable roof building from the front perspective views; (d)
illustration of DfFtR.
ISPRS Int. J. Geo-Inf. 2021, 10, 650 8 of 24
(a) (b)
Figure 5. (a) The main ridge of a flush gable roof building; (b) the roof point cloud density distribu-
tion on the XOY plane.
3. Feature Extraction
The input data are an entire architecture point cloud. Considering the calculation
process of the selected features, the roof and the ridges should be extracted before gener-
ating the features. In this section, a roof extraction method based on the change of projec-
tive areas and a ridge extraction method using the section lines are illustrated at first. On
the basis of the extracted roof and ridges, the features are generated in Section 3.3.
3.1. The Roof Extraction Method Based on the Change of Projective Areas
The shape of the Ming and Qing Dynasty official-style architecture roof is sloping or
curved and the roof eaves extend out of the beam frame. This results in the projective area
of the points becoming small from the roof eaves to the top of the roof on the XOY plane.
Based on the change of projective areas, a roof extraction method is proposed. The details
of this method are described as follows:
Suppose that the original point cloud represented the Ming and Qing Dynasty offi-
cial-style architecture (Figure 6a) is defined as = { ( )| = 1,2, ⋯ , },
is the number of points in . The point ( ) and the point ( )
belonging to the point cloud are the highest and lowest points along direction
separately.
Along the z direction, divide the point cloud into the several of subsets with inter-
val as is shown in Figure 6b. The interval is set as 0.1 m based on experience.
( )
The sampled points are defined as = = 1,2, ⋯ , , = int +1 . A
point from the point cloud is categorized to the subset which should meet
Equation (1).
∈
( − ) (1)
j = int +1
From 1 to M, wipe off the point set from the original point cloud in turn. After
each point subset is removed, project the remaining points onto the XOY planar co-
ordinates with a scale . If the number of points falls in a grid beyond 0, this grid is
ISPRS Int. J. Geo-Inf. 2021, 10, 650 9 of 24
3500
the first roof eave
3000
1500
1000
500
0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67 69
(a) (b)
(c) (d)
(e) (f)
Figure 7. The process of extracting ridges: (a) the top view of the original point cloud; (b) section
lines; (c) the points from section line along LX1 labeled as in (b); (d) the points from different section
lines along X axis; (e) the points from section line along LY1 labeled as in (b); (f) the extracted ridge
points.
3. SoRs generation
Project the ridge points onto the XOY plane. If the number of the points located in a
grid is beyond 1, the grid is labeled as 255. After these steps, the generated ridge lines still
have a width of 2–3 pixels. Subsequently, a Skeletonization algorithm [67] is performed to
thin the ridge lines. Finally, a straight-line detector based on the Freeman chain code [68]
is used to generate the ridge lines.
4. DoPP generation
The main ridge is parallel to the main direction of the architecture, and is located in
the middle area of the architecture. Based on this, the ridge line which represents the main
ridge is selected. Around the selected ridge line, create a buffer region. The number of
points located in this buffer region divided by the area of this buffer region is the DoPP.
Table 2. The attributed relational graph template of the ridges from pyramidal roofs.
= , , ,
= 〈 〉, 〈 〉, 〈 〉, 〈 〉, 〈 〉, 〈 〉
={ | , = 1,2, ⋯ , }
={ | , = 1,2, ⋯ , }
= {α = ⁄ }
= {θ = cos ( ) }
gable and hip roof is grouped into the flush gable roof category, overhanging gable
roof category or gable and hip roof category; otherwise, the unclassified roof is cate-
gorized as an overhanging gable round ridge roof, flush gable round ridge roof or
gable and hip round ridge roof.
Figure 9. Proposed hierarchical semantic network for the Ming and Qing official-style architecture
classification.
The doukou has 11 sizes or grades and the minimum dimensions of doukou is 1 cun
(3.5 cm) as is show in Figure 10. However, the eave column diameter D does not have an
exact value. To overcome this limitation, the corresponding sizes of the same components
from the wooden frame architecture with or without dougongs is applied to obtain an
approximate value of D. Based on this, D ranges from 3.6 doukou to 5.5 doukou. Consid-
ering the different grades of Ming and Qing official-style buildings, HoMRR ranges from
1 cun × 8 to 6 cun × 12 and CL ranges from 1 cun × 3.3 × 3.6 to 6 cun × 12.
ℎ should be equal to CL. The higher the grade of the Ming and Qing official-
style building is, the bigger CL is. Hence, ℎ should be higher than the minimum
CL-1 cun × 8. For ℎ , suppose that the density of point cloud is . In theory,
ℎ should be equal to × (2 × × + × )⁄( × ) when is
bigger than the width of the main ridge; otherwise, ℎ is equal to ×
( × + × )⁄( × ). In this paper, is user-defined. Here, we set as
. We can set this value to be the ℎ which can be set as 2 .
6. Performance Evaluation
6.1. Experimental Data Description
Many Ming and Qing official-style buildings have been established as major histori-
cal and cultural sites protected at the national level in China and are managed by corre-
sponding institutions. When we capture the point cloud, the authorization from these in-
stitutions is necessary. This results in the acquisition of the point cloud becoming very
difficult. In our experiments, three datasets were used to evaluate the performance of our
proposed method. The first two datasets consisted of point clouds captured by the terres-
trial laser scanning system or generated by UAV images and contained two roof types. To
test more roof types, we introduced a third dataset composed of the point clouds derived
from 3DsMAX models. The 3D information of these models is consistent with that of real
buildings and is reliable test data through inspection and comparison by experts. Each
dataset is described as follows:
The first dataset contained the point cloud of the Gate of Supreme Harmony and the
Hall of Complete Harmony labeled as rectangle 1 and 2 in Figure 11a. These point
cloud was captured by the terrestrial laser scanning (TLS) system. Figure 12 shows
the point cloud after registration in the commercial software package Leica Cyclone.
ISPRS Int. J. Geo-Inf. 2021, 10, 650 15 of 24
The point cloud density of the Gate of Supreme Harmony and the Hall of Complete
Harmony was 44,083 points/m2 and 43,416 points/m2, respectively.
(a) (b)
Figure 11. (a) Forbidden City; (b) Qutan Temple.
Figure 12. The point cloud of the Gate of Supreme Harmony and the Hall of Complete Harmony.
The second dataset was composed of the dense image matching (DIM) point cloud
of BaoGuang Hall located in Qutan Temple, QingHai province, China as is shown in
Figure 11b. 261 UAV images are collected by DJI Phantom4 which was composed of
a FC6310R camera with a 13.2 × 8.8 mm2 sensor size and a 2.41 μm pixel size. The
flight path surrounded the building as is shown in Figure 13. The distance from the
exposure points to this building varied from 20 m to 85 m. Considering the 8.8 mm
focal length and the photographic distance, the ground sampling distance (GSD) for
all cameras ranged from 0.5 cm to 2 cm. Relying on the commercial software pack-
age Bentley, this DIM point cloud was generated. The density of the generated DIM
point cloud was 59,737 points/m .
ISPRS Int. J. Geo-Inf. 2021, 10, 650 16 of 24
camera position
Figure 13. The flight path.
The third dataset included 3DsMAX models of Meridian Gate, LiJing Xuan, Gate of
Lasting Happiness labeled as rectangle 4, 5 and 6 in Figure 11a and a 3D model ex-
ample from a website. To satisfy the data requirements, we converted these 3D mod-
els into 3D point cloud based on the commercial software CloudCompare. The den-
sity of the point cloud was 95 points/m2.
There are six types of Ming and Qing Dynasty official-style architecture roofs: dou-
ble-eave hip roof, double-eave gable and hip roof, flush gable roof, round ridge roof, over-
hanging gable roof and pyramidal roof. The roof type information of each test building is
described in Table 4.
Hall of Complete
pyramidal roof
Harmony
double-eave ga-
BaoGuang Hall
ble and hip roof
double-eave hip
Meridian Gate
roof
ISPRS Int. J. Geo-Inf. 2021, 10, 650 17 of 24
overhanging ga-
A 3D model example
ble roof
of Lasting Happiness was categorized as a flush gable roof with round ridge and
other unclassified roofs were classified correctly.
The experimental results at each stage were consistent with the theoretical results
and all the test architecture roofs were classified into the correct categories. This shows
that our proposed method could get a good performance for the Ming and Qing Dynasty
official-style architecture roof classification.
Original point
cloud of test
800
areas on the X
pixels
15,000 400
1,500 3,000
10,000 20,000
tions 1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89
Extracted roof
Extracted ridge
points
double-eave gable and hip double-eave gable and hip flush gable roof with overhanging flush gable
Roof type pyramidal roof double-eave hip roof flush gable roof
roof roof round ridge roof
ISPRS Int. J. Geo-Inf. 2021, 10, 650 20 of 24
Table 6. The BaoGuang Hall roof recognition process based on point cloud with different density.
3,500
2,500
2,000
59,737 1,500
1,000
500
0
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70
4,000
3,500
the projective areas in pixels
3,000
2,500
2,000
5,974 1,500
1,000
500
0
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70
4,000
3,500
the projective areas in pixels
3,000
2,500
2,000
597 1,500
1,000
500
0
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70
Table 7. The Hall of Complete Harmony roof reorganization process based on point cloud with different density.
15,000
10,000
pixels
43,416 5,000
null
0
1 4 7 1013161922252831343740434649525558616467707376
20,000
the projective areas in
15,000
10,000
pixels
4,342 5,000
null
0
1 4 7 1013161922252831343740434649525558616467707376
20,000
the projective areas in
15,000
10,000
pixels
434 5,000
null
0
1 4 7 1013161922252831343740434649525558616467707376
Table 8. The Gate of Lasting Happiness roof reorganization process based on point cloud with different density.
7,000
6,000
he projective areas in pixels
5,000
4,000
95 3,000
2,000
1,000
0
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64
8,000
7,000
6,000
he projective areas in pixels
5,000
4,000
9.5 3,000
2,000
1,000
0
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64
8,000
7,000
6,000
4,000
1 3,000
2,000
1,000
0
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64
For the main ridge detection, under different point cloud density conditions, the
main ridges of BaoGuang Hall were detected as is shown in the fourth column of Table 6.
The distribution of the point cloud density on the XOY plane as is shown in the fourth
column of Table 8 illustrated that Gate of Lasting Happiness did not have a main ridge.
The experimental results were consistent with the true roof types. Notably, when the dis-
tance between points was greater than the height or width of the main ridge, our proposed
method may not have detected the main ridge. This is mainly because the points located
on the surface of the main ridge did not exist. This is an extreme case. In this paper, we
suppose that the point cloud can represent the details of the Ming and Qing official-style
architecture. Based on the above analysis, our proposed method is robust to the test point
clouds with different densities.
Moreover, the gaps, noise and occlusion of point clouds also influenced the experi-
mental results. In order to ensure that the point cloud data can successfully complete the
3D modeling, the quality of the collected point clouds should meet a certain criterion in
practical engineering. Hence, we can ignore these factors in our experiments.
Author Contributions: Conceptualization, Youqiang Dong and Miaole Hou; methodology, Biao Xu;
software, Youqiang Dong; validation, Miaole Hou, Yuhang Ji and Yihao Li; formal analysis, Biao
Xu; investigation, Yuhang Ji; resources, Miaole Hou data curation, Yuhang Ji; writing—original
ISPRS Int. J. Geo-Inf. 2021, 10, 650 22 of 24
draft preparation, Yuhang Ji and Yihao Li; writing—review and editing, Youqiang Dong; supervi-
sion, Biao Xu and Yihao Li; project administration, Miaole Hou; funding acquisition, Youqiang
Dong. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by National Key Research and Development Program of
China(2019YFC1520800); Scientific Research Project of Beijing Educational Committee
(KM202110016005); Beijing Postdoctoral Research Foundation (21009920004); Beijing Municipal Ed-
ucation Commission (KZ202110016021); Young Beijing Scholars Project (04147521504).
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.
References
1. Zhang, D. Cultural Symbols in Chinese Architecture. Archit. Des. Rev. 2019, 1, 2–17, doi:10.24294/adr.v1i1.556.
2. Armani, S.; Arbi, E. A Comparative Study on Chinese Architecture in Peninsular Malaysia and Mainland China. J. Des. Built
Environ. 2014, 14, 1.
3. Hu, Q.; Wang, S.; Fu, C.; Ai, M.; Yu, D.; Wang, W. Fine Surveying and 3D Modeling Approach for Wooden Ancient Architecture
via Multiple Laser Scanner Integration. Remote Sens. 2016, 8, 270, doi.org/10.3390/rs8040270.
4. Gomes, L.; Bellon, O.R.P.; Silva, L. 3D reconstruction methods for digital preservation of cultural heritage: A survey. Pattern
Recognit. Lett. 2014, 50, 3, doi:10.1016/j.patrec.2014.03.023.
5. Hu, Z.; Qin, X. Extended interactive and procedural modeling method for ancient Chinese architecture. Multimed. Tools Appl.
2020, 80, 5773–5807, doi:10.1007/s11042-020-09744-2.
6. Liu, J.; Wu, Z.-K. Rule-Based Generation of Ancient Chinese Architecture from the Song Dynasty. J. Comput. Cult. Herit. 2015,
9, 1–22, doi:10.1145/2835495.
7. Yang, X.; Grussenmeyer, P.; Koehl, M.; Macher, H.; Murtiyoso, A.; Landes, T. Review of built heritage modelling: Integration
of HBIM and other information techniques. J. Cult. Herit. 2020, 46, 350–360, doi:10.1016/j.culher.2020.05.008.
8. Calin, M.; Damian, G.; Popescu, T.; Manea, R.; Erghelegiu, B.; Salagean, T. 3D modeling for digital preservation of Romanian
heritage monuments. Agric. Agric. Sci. Procedia 2015, 6, 421–428.
9. Poux, F.; Billen, R.; Kasprzyk, J.-P.; Lefebvre, P.-H.; Hallot, P. A Built Heritage Information System Based on Point Cloud Data:
HIS-PC. ISPRS Int. J. Geo-Inf. 2020, 9, 588, doi:10.3390/ijgi9100588.
10. Kada, M.; McKinley, L. 3D building reconstruction from LiDAR based on a cell decomposition approach. Int. Arch. Photogramm.
Remote Sens. Spat. Inf. Sci. 2009, 38, W4.
11. Maas, H.G.; Vosselman, G. Two algorithms for extracting building models from raw laser altimetry data. ISPRS J. Photogramm.
Remote Sens. 1999, 54, 153–163, doi:10.1016/S0924-2716(99)00004-0.
12. Henn, A.; Gröger, G.; Stroh, V.; Plümer, L. Model driven reconstruction of roofs from sparse LIDAR point cloud. ISPRS J. Pho-
togramm. Remote Sens. 2013, 76, 17–29, doi:10.1016/j.isprsjprs.2012.11.004.
13. Zheng, Y.; Weng, Q. Model-driven reconstruction of 3D buildings using LiDAR data. IEEE Geosci. Remote Sens. Lett. 2015, 12,
1541–1545, doi:10.1109/LGRS.2015.2412535.
14. Vallet, B.; Pierrot-Deseilligny, M.; Boldo, D.; Brédif, M. Building footprint database improvement for 3D reconstruction: A split
and merge approach and its evaluation. ISPRS J. Photogramm. Remote Sens. 2011, 66, 732–742, doi:10.1016/j.isprsjprs.2011.06.005.
15. Lafarge, F.; Descombes, X.; Zerubia, J.; Pierrot-Deseilligny, M. Automatic building extraction from DEMs using an object ap-
proach and application to the 3D-city modeling. ISPRS J. Photogramm. Remote Sens. 2008, 63, 365–381, doi:10.1016/j.is-
prsjprs.2007.09.003.
16. Li, J. (Song Dynasty). Yingzao Fashi. Dongjing, Song Dynasty of China. 1103.
17. Qing Department of Qing Dynasty. Qing Gong Bu Gongcheng Zuofa Zeli. Beijing, Qing Dynasty of China. 1733.
18. Shen, Y.; Zhang, E.; Feng, Y.; Liu, S.; Wang, J. Parameterizing the Curvilinear Roofs of Traditional Chinese Architecture. Nexus
Netw. J. 2020, 23, 475–492, doi:10.1007/s00004-020-00512-1.
19. Liu, J. Component-driven procedural modeling for ancient Chinese architecture of the Qing Dynasty. Int. J. Archit. Herit. 2017,
12, 280–307, doi:10.1080/15583058.2017.1410253.
20. Li, L.; Tang, L.; Zhu, H.; Zhang, H.; Yang, F.; Qin, W. Semantic 3D Modeling Based on CityGML for Ancient Chinese-Style
Architectural Roofs of Digital Heritage. ISPRS Int. J. Geo-Inf. 2017, 6, 132, doi:10.3390/ijgi6050132.
21. Rahmatabadi, S.; Toushmalani, R. Physical order and disorder in Chinese architecture style. Aust. J. Basic Appl. Sci. 2011, 5, 1561–1565.
22. Kushwaha, S.K.P.; Yogender, Y.; Sara, R. A semi-automatic approach for roof-top extraction and classification from airborne
lidar. In Proceedings of the Seventh International Conference on Remote Sensing and Geoinformation of the Environment
(RSCy2019), Paphos, Cyprus, 18–21 March 2019, doi:10.1117/12.2532044.
23. Mohajeri, N.; Assouline, D.; Guiboud, B.; Bill, A.; Gudmundsson, A.; Scartezzini, J.L. A city-scale roof shape classification using
machine learning for solar energy applications. Renew. Energy 2018, 121, 81–93.
ISPRS Int. J. Geo-Inf. 2021, 10, 650 23 of 24
24. Zang, A.; Zhang, X.; Chen, X.; Agam, G. Learning-based roof style classification in 2D satellite images. Proc. SPIE 2015, 9473,
doi:10.1117/12.2180393.
25. Assouline, D.; Mohajeri, N.; Scartezzini, J.L. Building rooftop classification using random forests for large-scale PV deployment.
In Proceedings of the Earth Resources and Environmental Remote Sensing/GIS Applications VIII, Warsaw, Poland, 5 October
2017; Volume 10428, p. 1042806.
26. Aissou, B.E.; Aissa, A.B.; Dairi, A.; Harrou, F.; Wichmann, A.; Kada, M. Building Roof Superstructures Classification from Im-
balanced and Low Density Airborne LiDAR Point Cloud. IEEE Sens. J. 2021, 21, 14960–14976.
27. Zhang, X.; Zang, A.; Agam, G.; Chen, X. Learning from synthetic models for roof style classification in point cloud. In Proceed-
ings of the 22nd ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems—SIGSPATIAL,
Dallas, TX, USA, 4–7 November 2014; pp. 263–270, doi:10.1145/2666310.2666407.
28. Axelsson, M.; Soderman, U.; Berg, A.; Lithen, T. Roof Type Classification Using Deep Convolutional Neural Networks on Low
Resolution Photogrammetric Point Clouds from Aerial Imagery. In Proceedings of the 2018 IEEE International Conference on
Acoustics, Speech and Signal Processing (ICASSP), Calgary, AB, Canada, 15–20 April 2018; pp. 1293–1297,
doi:10.1109/ICASSP.2018.8461740.
29. Partovi, T.; Fraundorfer, F.; Azimi, S.; Marmanis, D.; Reinartz, P. Roof type selection based on patch-based classification using
deep learning for high resolution satellite imagery. ISPRS-Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. 2017, XLII-1/W1,
653–657.
30. Castagno, J.; Atkins, E. Roof Shape Classification from LiDAR and Satellite Image Data Fusion Using Supervised Learning.
Sensors 2018, 18, 3960, doi:10.3390/s18113960.
31. Bittner, K.; Körner, M.; Fraundorfer, F.; Reinartz, P. Multi-Task cGAN for Simultaneous Spaceborne DSM Refinement and Roof-
Type Classification. Remote Sens. 2019, 11, 1262.
32. Buyukdemircioglu, M.; Can, R.; Kocaman, S. Deep learning based roof type classification using very high resolution aerial
imagery. Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci. 2021, XLIII-B3, 55–60, doi:10.5194/isprs-archives-XLIII-B3-2021-55-
2021, 2021.
33. Grilli, E.; Özdemir, E.; Remondino, F. Application of machine and deep learning strategies for the classification of heritage point
clouds. Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. 2019, XLII-4/W18, 447–454, doi:10.5194/isprs-archives-XLII-4-W18-447-
2019.
34. Tian, Y.; Song, W.; Sun, S.; Fong, S.; Zou, S. 3D object recognition method with multiple feature extraction from LiDAR point
clouds. J. Supercomput. 2019, 75, 4430–4442, doi:10.1007/s11227-019-02830-9.
35. Yu, Z. Intrinsic shape signatures: A shape descriptor for 3D object recognition. In Proceedings of the 2009 IEEE 12th Interna-
tional Conference on Computer Vision Workshops, ICCV Workshops, Kyoto, Japan, 27 September–4 October 2009.
36. Mian, A.; Bennamoun, M.; Owens, R. On the Repeatability and Quality of Key points for Local Feature-based 3D Object Re-
trieval from Cluttered Scenes. Int. J. Comput. Vis. 2010, 89, 348–361, doi:10.1007/s11263-009-0296-z.
37. Hui, C.; Bhanu, B. 3D free-form object recognition in range images using local surface patches. Pattern Recognit. Lett. 2007, 28,
1252–1262, doi:10.1016/j.patrec.2007.02.009.
38. Rusu, R.B.; Blodow, N.; Beetz, M. Fast Point Feature Histograms (FPFH) for 3D registration. In Proceedings of the 2009 IEEE
International Conference on Robotics and Automation, Kobe Japan, 12–17 May 2009; pp. 3212–3217, doi:10.1109/RO-
BOT.2009.5152473.
39. Guo, Y.; Sohel, F.; Bennamoun, M.; Lu, M.; Wan, J. Rotational projection statistics for 3D local surface description and object
recognition. Int. J. Comput. Vis. 2013, 105, 63, doi:10.1007/s11263-013-0627-y.
40. Tombari, F.; Salti, S.; Di Stefano, L. Unique Signatures of Histograms for Local Surface Description. In Proceedings of the 11th
European Conference on Computer Vision Conference on Computer Vision: Part III, Crete, Greece, 5–11 September 2010;
Springer: Berlin/Heidelberg, Germany, 2010; pp. 356–369, doi:10.1007/978-3-642-15558-1_26.
41. Mikolajczyk, K.; Schmid, C. A performance evaluation of local descriptors. IEEE Trans. Pattern Anal. Mach. Intell. 2005, 27, 1615–1630,
doi:10.1109/TPAMI.2005.188.
42. Rusu, R.B.; Holzbach, A.; Beetz, M.; Bradski, G. Detecting and segmenting objects for mobile manipulation. In Proceedings of
2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops, Kyoto, Japan, 27 September–4
October 2009; IEEE: Kyoto, Japan, 2009.
43. Rusu, R.B.; Bradski, G.; Thibaux, R.; Hsu, J. Fast 3D recognition and pose using the Viewpoint Feature Histogram. In Proceed-
ings of the 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems, Taipei, Taiwan, 18–22 October 2010;
IEEE: Taipei, Taiwan, 2010.
44. Aldoma, A.; Vincze, M.; Blodow, N.; Gossow, D.; Gedikli, S.; Rusu, R.B.; Bradski, G. CAD-model recognition and 6D OF pose
estimation using 3D cues. In Proceedings of the IEEE International Conference on Computer Vision Workshops, ICCV 2011
Workshops, Barcelona, Spain, 6–3 November 2011; IEEE: Barcelona, Spain, 2011.
45. Wohlkinger, W.; Vincze, M. Ensemble of shape functions for 3D object classification. In Proceedings of the 2011 IEEE Interna-
tional Conference on Robotics and Biomimetics, Phuket, Thailand, 7–11 December 2012; IEEE: Phuket, Thailand, 2012.
46. Schnabel, R.; Wahl, R.; Wessel, R.; Klein, R. Shape Recognition in 3D Point-Clouds. In Proceedings of the 16th International
Conference in Central Europe on Computer Graphics, Visualization and Computer Vision’2008, Plzen-Bory, Czech Republic,
4–7 February 2008.
ISPRS Int. J. Geo-Inf. 2021, 10, 650 24 of 24
47. Cheng, Y.-M.; Ding, H.-X.; Wang, Y.-X.; Zhang, H.-H. Curved Object Recognition Based on Geometrical Features. J. Image Graph.
2000, 5, 573–579.
48. Hao, W.; Wang, Y. Structure-based object detection from scene point clouds. Neurocomputing 2016, 191, 148, doi:10.1016/j.neu-
com.2015.12.101.
49. Berner, A.; Li, J.; Holz, D.; Stuckler, J.; Behnke, S.; Klein, R. Combining contour and shape primitives for object detection and
pose estimation of prefabricated parts. In Proceedings of the 2013 IEEE International Conference on Image Processing, Mel-
bourne, Australia, 15–18 September 2013; p. 3326, doi:10.1109/icip.2013.6738685.
50. Dehbi, Y.; Henn, A.; Gröger, G.; Stroh, V.; Plümer, L. Robust and fast reconstruction of complex roofs with active sampling from
3D point clouds. Trans. GIS 2020, 12659, doi:10.1111/tgis.12659.
51. Zeybek, M. Classification of UAV point clouds by random forest machine learning algorithm. Turk. J. Eng. 2021, 5, 51–61,
doi:10.31127/tuje.669566.
52. Wang, Z.; Zhang, L.; Fang, T.; Mathiopoulos, P.T.; Tong, X.; Qu, H.; Xiao, Z.; Li, F.; Chen, D. A Multiscale and Hierarchical
Feature Extraction Method for Terrestrial Laser Scanning Point Cloud Classification. IEEE Trans. Geosci. Remote Sens. 2015, 53,
2409–2425, doi:10.1109/TGRS.2014.2359951.
53. Guo, B.; Huang, X.; Zhang, F.; Sohn, G. Classification of airborne laser scanning data using JointBoost. ISPRS J. Photogramm.
Remote Sens. 2015, 100, 71–83, doi:10.1016/j.isprsjprs.2014.04.015.
54. Yi, Z.; Wang, H.; Duan, G.; Wang, Z. An Airborne LiDAR Building-Extraction Method Based on the Naive Bayes-RANSAC
Method for Proportional Segmentation of Quantitative Features. J. Indian Soc. Remote Sens. 2020, 1–12, doi:10.1007/s12524-020-
01222-4.
55. Eckart, B.; Kelly, A. REM-Seg: A robust EM algorithm for parallel segmentation and registration of point clouds. In Proceedings
of the 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2013), Tokyo, Japan, 3–7 November
2013; pp. 4355–4362, doi:10.1109/IROS.2013.6696981.
56. Qi, C.R.; Su, H.; Mo, K.; Guibas, L.J. Pointnet: Deep learning on point sets for 3D classification and segmentation. In Proceedings
of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA, 21–26 July 2017; pp. 652–660.
57. Qi, C.R.; Yi, L.; Su, H.; Guibas, L.J. Pointnet++: Deep hierarchical feature learning on point sets in a metric space. arXiv 2017,
arXiv:1706.02413.
58. Li, Y.; Bu, R.; Sun, M.; Wu, W.; Di, X.; Chen, B. Pointcnn: Convolution on x-transformed points. arXiv 2018, arXiv:1801.07791.
59. Su, H.; Jampani, V.; Sun, D.; Maji, S.; Kalogerakis, E.; Yang, M.H.; Kautz, J. Splatnet: Sparse lattice networks for point cloud
processing. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, 18–
23 June 2018; pp. 2530–2539.
60. Wang, Y.; Sun, Y.; Liu, Z.; Sarma, S.E.; Bronstein, M.M.; Solomon, J.M. Dynamic graph cnn for learning on point clouds. ACM
Trans. Graph. 2019, 38, 146.
61. Wu, Z.; Song, S.; Khosla, A.; Yu, F.; Zhang, L.; Tang, X.; Xiao, J. 3D shapenets: A deep representation for volumetric shapes. In
Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA, 7–12 June 2015; pp. 1912–1920.
62. Geiger, A.; Lenz, P.; Stiller, C.; Urtasun, R. Vision meets Robotics: The KITTI Dataset. Int. J. Robot. Res. IJRR 2013, 32, 1231–1237.
63. De Deuge, M.; Quadros, A.; Hung, C.; Douillard, B. Unsupervised feature learning for classification of outdoor 3D scans. In
Proceedings of the Australasian Conference on Robitics and Automation, Sydney, Australia, 2–4 December 2013; Volume 2, p. 1.
64. Hackel, T.; Savinov, N.; Ladicky, L.; Wegner, J.D.; Schindler, K.; Pollefeys, M. SEMANTIC3D.NET: A new large-scale point
cloud classification benchmark. ISPRS Ann. Photogram. Remote Sens. Spat. Inf. Sci. 2017, IV-1-W1, 91–98.
65. Armeni, I.; Sener, O.; Zamir, A.R.; Jiang, H.; Brilakis, I.; Fischer, M.; Savarese, S. 3D semantic parsing of large-scale indoor spaces.
In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 27–30 June 2016; pp.
1534–1543.
66. Pierdicca, R.; Paolanti, M.; Matrone, F.; Martini, M.; Morbidoni, C.; Malinverni, E.S.; Frontoni, E.; Lingua, A.M. Point Cloud
Semantic Segmentation Using a Deep Learning Framework for Cultural Heritage. Remote Sens. 2020, 12, 1005.
67. Iwanowski, M.; Soille, P. Fast Algorithm for Order Independent Binary Homotopic Thinning. In Proceedings of the Interna-
tional Conference on Adaptive and Natural Computing Algorithms, Warsaw, Poland, 11–14 April 2007; Springer: Berlin, Ger-
many, 2007; pp. 606–615.
68. Lu, X.; Yao, J.; Li, K.; Li, L. CannyLines: A parameter-free line segment detector. In Proceedings of the 2015 IEEE International
Conference on Image Processing (ICIP), Quebec City, QC, Canada, 27–30 September 2015; pp. 507–511.
69. Čibej, U.; Mihelič, J. Improvements to Ullmann’s Algorithm for the Subgraph Isomorphism Problem. Int. J. Pattern Recognit.
Artif. Intell. 2015, 29, 1550025, doi:10.1142/S0218001415500251.
70. Dong, Y.; Zhang, L.; Cui, X.; Ai, H.; Xu, B. Extraction of Buildings from Multiple-View Aerial Images Using a Feature-Level-
Fusion Strategy. Remote Sens. 2018, 10, 1947, doi:10.3390/rs10121947.
71. Huo, P.; Hou, M.; Dong, Y.; Li, A.; Ji, Y.; Li, S. A Method for 3D Reconstruction of the Ming and Qing Official-Style Roof Using
a Decorative Components Template Library. ISPRS Int. J. Geo-Inf. 2020, 9, 570, doi:10.3390/ijgi9100570.