3D城市数据库：CityGML标准应用

3星 · 超过75%的资源需积分: 50 180 浏览量更新于2024-07-19 1 收藏 46.82MB PDF 举报

"3D City Database是针对CityGML标准的一种3D地理数据库，用于存储和管理城市环境的三维模型数据。该资源可能是CityGML的第三版3.3.0版本的相关文档，由多个机构和专家共同开发，包括德国慕尼黑工业大学的摄影测量与遥感及制图系、法兰克福市测绘局、IDAC Ltd（英国）、virtualcitySYSTEMS GmbH（柏林）以及慕尼黑工业大学的地学信息学系等。封面图像由这些机构提供，其中包含柏林的3D城市模型数据。" CityGML（城市地理标记语言）是一种开放的、基于XML的数据模型和交换格式，专门用于描述城市的三维几何、拓扑、语义和功能特性。它允许城市对象如建筑物、道路、绿地等以标准化的方式进行建模，便于在不同的应用程序之间共享和交换城市数据。 3D City Database是专门为CityGML设计的，它提供了对3D城市模型的有效存储和查询机制。这个数据库系统能够存储CityGML的不同级别抽象（LOD，Level of Detail），从简单的几何表示到复杂的、富含语义的模型。通过3D Geodatabase，用户可以执行空间分析、场景可视化、城市规划等多种任务，这对于城市管理和决策支持至关重要。在文档中提到的参与者列表，包括Thomas H. Kolbe、Zhihang Yao、Claus Nagel、Richard Redweik、Philipp Willkomm、György Hudra和Arda Müftüoglu等人，他们在开发3D City Database方面发挥了积极作用。他们的电子邮件地址被提及，这可能意味着他们为用户提供技术支持或合作机会。此外，资源还提到了一些机构，如M.O.S.S. Computer Grafik Systeme GmbH，它们可能提供了技术贡献或软件解决方案。这表明3D City Database不仅是一个理论概念，而是由业界专家和公司合作实现的实际产品，具有实际应用价值。 3D City Database for CityGML是一个强大且标准化的工具，用于处理和管理城市环境的复杂三维数据，结合CityGML的标准化框架，使得跨组织、跨系统的城市数据交换成为可能，对于城市规划、环境模拟、灾害响应等多个领域具有重要意义。

16 3D Geodatabase for CityGML 2016

WebGL Virtual Globe. A tiling strategy is supported where only tiles in the vicinity of

the viewer’s location are being loaded facilitating the visualization of even very large

3D city and landscape models. Information balloons for all objects can be configured

by the user.

 Tool for exporting data to spreadsheets: The ‘Spreadsheet Generator’ (SPSHG)

allows exporting thematic data of 3D objects into tables in CSV and Microsoft Excel

format which can be uploaded to a Google Spreadsheet within the Google Document

Cloud. For every selected geoobject one row is being exported where the first column

always contains the GMLID value of the respective object. The further columns can

be selected by the user. This tool can be used to export attribute data from e.g.

buildings like the class, function, usage, roof type, address, and further generic

attributes that may contain information like the building energy demand, potential

solar energy gain, noise level on the facades etc. The spreadsheet rows can be linked

to the visualization model generated by the KML/COLLADA/glTF Exporter. This is

shown in Appendix B.

 Tool for 3D visualization and interactive exploration of 3D models on the web:

The ‘3DCityDB-Web-Map-Client’ is a WebGL-based 3D web viewer which extends

the Cesium Virtual Globe to support efficient displaying, caching, prefetching,

dynamic loading and unloading of arbitrarily large pre-styled 3D visualization models

in the form of tiled KML/glTF datasets generated by the KML/COLLADA/glTF

Exporter. It provides an intuitive user interface to facilitate rich interaction with 3D

visualization models by means of the enhanced functionalities like highlighting the

objects of interests on mouseover and mouseclick as well as hiding, showing, and

shadowing them. Moreover, the 3DCityDB-Web-Map-Client is able to link the 3D

visualization model with an online spreadsheet (Google Fusion Table) in the Google

Cloud and allows viewing and querying the thematic data of every city object

according to its GMLID.

 Web Feature Service (WFS) 2.0: The 3DCityDB comes with an OGC compliant

implementation of a basic WFS 2.0 allowing web-based access to the 3D city objects

stored in the database. WFS clients can directly connect to this interface and retrieve

3D content for a wide variety of purposes. The implementation currently satisfies the

Simple WFS conformance class. An implementation of a full, transactional WFS is

commercially available from one of the development partners, see Appendix C.

 Support of different kinds of multi-representations: Levels of detail, different

appearances, (and with Oracle RDBMS only) planning versions and history:

Every geoobject as well as the DTM can be represented in five different resolution or

fidelity steps (Levels of Detail, LOD). With increasing LOD, objects do not only

obtain a more precise and finer geometry, but do also gain a thematic refinement.

Different appearance data may be stored for each city object. Appearance relates to

any surface-based theme, e.g. infrared radiation or noise pollution, not just visual

3D Geodatabase for CityGML 2016 17

properties. Consequently, data provided by appearances can be used as input for both

presentation and analysis of virtual 3D city models. The database supports feature

appearances for an arbitrary number of themes per city model. Each LOD of a feature

can have individual appearances. Appearances can represent – among others – textures

and georeferenced textures. All texture images can be stored in the database.

The version and history management employs Oracle’s Workspace Manager and,

hence, is only available for 3DCityDB instances running on an Oracle RDBMS. It is

largely transparent to application programs that work with the database. Procedures

saved within the database (Stored Procedures) are provided, which allow for the

management of planning alternatives and versions via application programs.

 Complex digital terrain models: DTMs may be represented in four different ways in

CityGML and therefore also in the 3D city database: regular grids, triangular irregular

networks (TINs), 3D mass points and 3D break lines. For every level of detail, a

complex DTM consisting of any number of DTM components and DTM types can be

defined. Besides, it is possible to combine certain kinds of DTM representations for

the same geographic area with each other (e.g. mass points and break lines or grids

and break lines). In Oracle Spatial (but not Locator) Grid-based DTMs may be of

arbitrary size and are composed from separate tiles to a single overall grid using the

Oracle GeoRaster functionality. Please note that the Import/Export tool provides

functions to read and write TIN, mass point, and break line DTM components, but not

for raster based DTMs. GeoRaster data would have to be imported and exported using

other tools from e.g. Oracle, ESRI, or Safe Software.

 Complex city object modelling: The representation of city objects in the 3D city

database ranges from coarse models to geometrically and semantically fine grained

structures. The underlying data model is a complete realization of the CityGML data

model for the levels of detail (LOD) 0 to 4. For example, buildings can be represented

by simple, monolithic objects or can consist of an aggregation of building parts.

Extensions of buildings, like balconies and stairs, can be classified thematically and

provided with attributes just as single surfaces can be. LOD4 completes a LOD3

model by adding interior structures for 3D objects. For example, LOD4 buildings are

composed of rooms, interior doors, stairs, and furniture. This allows among other

things to select the floor space of a building, so that it can later be used e.g. to derive

SmartBuildings or to form 3D solids by extrusion [Döllner et al. 2005]. Buildings can

be assigned addresses that are also stored in the 3D city database. Their implemen-

tation refers to the OASIS xAL Standard, which maps the address formats of the

different countries into a unified XML schema. In order to model whole complexes of

buildings, single buildings can be aggregated to form special building groups. The

same complex modelling applies to the other CityGML feature types like bridges,

tunnels, transportation and vegetation objects, and water bodies.

 Representation of generic and prototypical 3D objects: Generic objects enable the

storage and management of 3D geoobjects that are not explicitly modelled in

18 3D Geodatabase for CityGML 2016

CityGML yet, for example dams or city walls, or that are available in a proprietary file

format only. This way, files from other software systems like architecture or computer

graphics programs can be imported directly into the database (without interpretation).

However, application systems that would like to use these data must be able to

interpret the corresponding file formats after retrieving them back from the 3D

geodatabase.

Prototypical objects are used for memory-efficient management of objects that occur

frequently in the city model and that do not differ with respect to geometry and

appearance. Examples are elements of street furniture like lanterns, road signs or

benches as well as vegetation objects like shrubs, certain tree types etc. Every instance

of a prototypical object is represented by a reference to the prototype, a base point and

a transformation matrix for scaling, rotating and translating the prototype.

The geometries (and appearances like textures, colors etc.) of generic objects as well

as prototypes can be stored either using the geometry datatype of the spatial database

management system (Oracle Spatial/Locator or PostGIS) or in proprietary file formats.

In the latter case a single file may be saved for every object, but the file type (MIME

type), the coordinate transformation matrix that is needed to integrate the object into

the world coordinate reference system (CRS) as well as the target CRS have to be

specified.

 Extendable object attribution: All objects in the 3D geodatabase can be augmented

with an arbitrary number of additional generic attributes. This way, it is possible to

add further thematic information as well as further spatial properties to the objects at

any time. In combination with the concept of generic 3D objects this provides a highly

flexible storage option for object types which are not explicitly defined in the

CityGML standard. Every generic attribute consists of a triple of attribute name, data

type, and value. Supported data types are: string; integer and floating-point numbers;

date; time; binary object (BLOB, e.g. for storing a file); geometry object according to

the specific geometry data type of Oracle or PostGIS respectively; simple, composite,

or aggregate 3D solids or surfaces. Please note that generic attributes of type BLOB or

geometry are not allowed as generic attributes in CityGML (and will, thus, not be

exported by the CityGML exporter). However, it may be useful to store binary data

associated with the individual city objects, for example, to store derived 3D computer

graphics representations.

 Free, also recursive grouping of geoobjects: Geoobjects can be grouped arbitrarily.

The aggregates can be named and may also be provided with an arbitrary number of

generic attributes (see above). Object groups may also contain object groups, which

leads to nested aggregations of arbitrary depth. In addition, for every object of an

aggregation, its role in the group can be specified explicitly (qualified association).

 External references for all geoobjects: All geoobjects can be provided with an

arbitrary number of references to corresponding objects in external data sources (i.e.

hyperlinks / linked data). For example, in case of building objects this allows to store

3D Geodatabase for CityGML 2016 19

e.g. the IDs of the corresponding objects in official cadasters, digital landscape

models, or Building Information Models (BIM). Each reference consists of an URI to

the external data store or database and the corresponding object ID or URI within that

external data store or database.

 Flexible 3D geometries: The geometry of most 3D objects can be represented through

the combination of solids and surfaces as well as any - also recursive - aggregation of

these elements. Each surface may has attached different textures and colors on both its

front and back face. It may also comprise information on transparency. Additional

geometry types (any geometry type supported by the spatial database management

system Oracle Spatial/Locator or PostGIS) can be added to the geoobjects by using

generic attributes.

 Open Source and Platform Independence: The entire software is freely accessible

to the interested public. The 3DCityDB is licensed under the Apache License, Version

2.0, which allows including 3DCityDB in commercial systems. You may obtain a

copy of the Apache License at http://www.apache.org/licenses/LICENSE-2.0. Both

the Importer/Exporter tool and the Web Feature Service are implemented in Java and

can be run on different platforms and operating systems.

1.2 System and design decisions

The 3D City Database is implemented as a relational database schema using the spatial

datatypes provided by a spatially-enhanced relational database management system

(SRDBMS). Above, external software applications and database stored procedures are

provided working on this database schema. Since only Oracle with the Spatial or Locator

licensing option (10G R2 or higher) and PostgreSQL (9.1 or higher) with PostGIS extension

(2.0 or higher) offer comprehensive support for 3D spatial data, the 3D City Database schema

is being provided for these two systems only.

In addition to the general advantages arising from the usage of a widely used relational

database management system (RDBMS), both Oracle Spatial/Locator and PostgreSQL/

PostGIS offer some important performance characteristics that allow an efficient implemen-

tation of the required functionalities:

 Both RDBMS support spatial data types with coordinates ranging from 2D to 4D.

Spatial indexes and filters can be 2D or 3D allowing for efficient spatial selections in

very large city models. Furthermore, the spatial data types are supported by a number

of commercial and Open Source GIS that provide a database connection as for

example ESRI’s ArcGIS/ArcSDE or Safe Software’s Feature Manipulation Engine

(FME). This enables such systems to directly access the data stored in the 3D

geodatabase.

 Rules can be implemented using stored procedures and trigger mechanisms which

propagate updates of objects to likewise affected objects in the database (transparent

for the user).

The da

oriente

excepti

In orde

done m

regardi

perfor

Surface

special

individ

assigne

explicit

stored i

on the

The pr

Java la

under

exploiti

1.3

1.3.1

Improv



New fe



a model o

data mod

n that geo

to achiev

anually wi

g the us

ance. Thes

ased geo

ay: they

al tuple i

multiple

y linkable

addition

like the c

vided soft

guage in

icrosoft

ng multi-t

ist of c

Notable

d and upd

pdated d

o Re-

3.0.

o Ad

ompletel

he Import

ome

ugf

tures and

Support fo

xport, the

irectory s

ewly incl

irect 3D v

nline spre

the 3D Ci

l of CityG

etry obje

e high per

h a numbe

ge of the

e are docu

etries lik

re decomp

one big

ppearance

to the corr

o their dec

mputation

are tools

rder to b

indows, L

reading on

hange

change

ted feature

tabase sch

dded GML

FACE_GE

0).

ed GMLID

evised an

/Export to

xes and im

unctionalit

glTF exp

generated

ructure acc

uded Cesi

sualizatio

dsheets) e

y Databas

L has be

ts are map

ormance f

of optimi

CityGML

ented in c

Polygons,

sed into th

urface tab

(e.g. text

sponding s

mposed b

f the volu

like the I

platform

nux, and

multiproce

betwe

s and funct

ma:

D_CODE

METRY,

and GML

updated

l and the

rovement

es:

rt in addit

ML/COL

rding to a

ased 3

and intera

ported usi

is based

n mapped

ed to the

r data ma

ations. A

concepts

apte

2.1.

TINs, Mul

eir primiti

e. The re

res) in Cit

rface. For

undary su

porter/Ex

ndependen

pple Mac

sor or mul

n soft

n 3.3.0

onalities:

PACE col

URFACE

ODE

ocumentat

FS interfa

of the Im

on to the

ADA/glT

specific til

web vie

tive explo

g the Imp

n the City

to a purely

patial data

ipulations

ew simpli

in the re

iSurfaces

e surfaces

son for th

GML and,

Solids als

faces allo

orter appli

. The tool

OS X. Hi

i-core CP

are v

nd 3.0.

mn to

_DATA,

PACE colu

on.

e now req

ort/Export

ML/COL

models ar

ng schema.

er (3DCit

ation of th

rter/Expor

eodatabase

ML 2.0 st

relational

pes provi

and querie

ying assu

l world

s well as S

nd each su

s is that e

thus, each

the solid g

ing to appl

ation are

have bee

h perform

systems.

rsions

YOBJECT

ITYMOD

n to

ire Java 8.

ool.

ADA exp

organize

DB-Web-

datasets (

er.

for CityG

andard. Th

ata model

ed by the

s the map

ptions wh

elping to

olids are st

rface is sto

ch surfac

ppearance

eometry o

y spatial o

mplement

confirme

nce is ach

APPEA

L (were dr

ESS table

rt. In cas

with a hie

ap-Clien

ML/glTF

L 2016

object-

with the

DBMS.

ing was

re made

increase

red in a

ed as an

can be

must be

ects are

erations

d in the

to run

eved by

ANCE,

pped in

of tiled

archical

) allows

odes +

剩余285页未读，继续阅读

chenwanpeng617

粉丝: 8
资源: 14

3D城市数据库：CityGML标准应用

柏林3D城市数据库：CityGML与CityDB的整合与应用

CityViz: 实现CityGML文件3D渲染的交互式WebGL应用

3Dtiles数据转换：CityGML到TilesetWithDiscreteLOD

3dcitydb, 3D 城市数据库开源CityGML数据库.zip

3dcitydb：3D城市数据库-开源CityGML数据库

3D城市数据库-开源CityGML数据库_PLpgSQL_PLSQL_下载.zip

citygml4hs:适用于CityGML的开源Haskell API。 根据CityGML规范自主开发-Source City

CityGML 2.0

cityGML 标准

citygml-to-3dtiles:从CityGML转换为Cesium 3D Tiles

最新资源

citygml4hs:适用于CityGML的开源Haskell API。根据CityGML规范自主开发-Source City