3D Modeling of Buildings

Outstanding Sites

Raphaële Héno, Laure Chandelier

• Publisher: Wiley-ISTE
• Year: 2014
• ISBN:
  9781118648902
• Edition: 1
• Topic: Technik & Maschinenbau
• Subtopic: Mikroelektronik

Conventional topographic databases, obtained by capture on aerial or spatial images provide a simplified 3D modeling of our urban environment, answering the needs of numerous applications (development, risk prevention, mobility management, etc.). However, when we have to represent and analyze more complex sites (monuments, civil engineering works, archeological sites, etc.), these models no longer suffice and other acquisition and processing means have to be implemented. This book focuses on the study of adapted lifting means for "notable buildings". The methods tackled in this book cover lasergrammetry and the current techniques of dense correlation based on images using conventional photogrammetry.

1

Specific Requirements for the 3D Digitization of Outstanding Sites

1.1. The current offer for high-resolution 3D data

In France, the IGN (Institut national de l’information géographique et forestière or the National Institute for Geographical and Forestry Information) produces and maintains the RGE¹ (Référentiel à Grande Echelle or Large-scale reference) over the entirety of the territory, containing four levels of information: a high-resolution orthoimage, an “addresses” database, a georeferenced cadastral map and a 3D topographic database of metric precision containing a description of elements of the landscape, classed according to type. Work is underway on a new-generation RGE, which is more precise and gives better resolution and detail, largely due to the cooperation of a number of stakeholders who collect and manage geographical information throughout the territory.

The IGN is also implementing a semi-automatic production process using high-resolution aerial images (pixel size below 20 cm), which are highly redundant, (significant overlap between images) leading to the production of vector databases in line with the LOD2 (level of detail) specifications of the CityGML² standards. It has been implemented in some large French cities, such as Paris, Nantes, Rennes, Besançon and Marseille. It shows the various districts, roof structures, vegetation and generalized street furniture. This product, Bati3D^® [MAI 04], not only contains a survey of the buildings’ forms (or groups of buildings if they are connected), but also their appearance, since the volume reproduced via automated photogrammetric processes is textured by aerial images. It is possible to improve the rendering and the precision of the façades and street furniture by adding data acquired in the field by vehicles equipped with stereoscopic cameras, lasers scanners and direct georeferencing devices (GNSS³ antenna, inertial navigational system or odometer). These devices are rarely self-sufficient in urban environments, but are essential for initiating the georeferencing of images through photogrammetry.

Many local authorities (well-known examples include Monaco⁴, Geneva⁵ and the urban community of Lyon whose model is now available to be consulted on-line) have financed digital 3D models of their territory, believing that the traditional products available were not able to meet their requirements. In general, specific teams manage the production of Digital Elevation Models (DEMs), orthoimages and vector models representing buildings, street furniture and vegetation, with the use of aerial images and adapted lidar technology. Large internet-based companies are also working on 3D representations of towns, offering navigation services in the most realistic reproductions of our environment. An example of this is Google, which offers detailed reproductions of entire districts. A number of different methods are used to achieve this from traditional photogrammetric stereoplotting, through to pure infographics and the production from multi-images available to every Internet user⁶. This range of processes gives rise to questions regarding traceability of data and the amount of confidence which can be invested in it.

The quality of aerial images and the ever more automated computerized tools allow for the large-scale production of urban databases with increasingly detailed resolutions. However, the majority of these databases solely aim to show towns from an external view-point, without looking into the buildings or going underground.

On the other hand, the architect’s digital model, or Building Information Modeling (BIM), is exhaustive, showing the building or the district from all angles: internal, external, underground and from above. Either this precedes the construction phase and gives an architect’s view, or it is used in urban restoration projects (modernizing old buildings), and it meets the requirement of providing a model “as-built”.

The methods of digitization discussed in this text provide a description of a particular building, which is more applicable than that given by other products available in terms of better spatial resolution or greater comprehensiveness. The text refrains from examining computer-aided design (CAD) processes, which concentrate more on the image than the accuracy of the survey. Here, the focus will be on geometric methods which, while keeping up-to-date, are able to remain objective and render a building as it really is and not as it may be imagined.

1.2. Statement of requirements

The digitization methods which are addressed in this text involve the following stages [BAR 11]:

– presentation of requirements;

– choice of techniques to be applied;

– data acquisition in the field;

– data processing;

– quality control;

– layout and delivery.

1.2.1. Potentials

Drawing up the statement of requirements for digitization, prior to the production of building specifications for the contractor, is a challenging exercise. A comprehensive, precise and well geolocalized digitization of an outstanding site may be requested in the case of an inventory, a restoration or an archeological dig, or with the aim to make it accessible to everyone everywhere with a digital model that can be consulted online (virtual tourism). Requirements regarding content, resolution, precision or up-to-dateness obviously vary depending on the context. However, the statement of requirements is often initially summarized in an idealistic manner: the 3D model of everything at the highest resolution. Budgetary constraints tend to bring project sponsors back down to earth. The reasons motivating digitization, in terms of potentials, should, therefore, be highlighted. The following examples can be given:

– Understanding the monument:

- establish a model of movements;

- recover the history of its construction;

- analyze its distortions.

– Providing a collection of measurements:

- count openings;

- measure the surface of the roofs;

- display different sections of the structure;

- study the stability of the monument.

– Locate the monument in time and space:

- superpose elevation to previous or future surveys;

- create an evolving 3D model;

- visualize the site and its positioning within its environment;

- view installations and existing equipment;

- show future projects or constructions.

– Communicate, broadcast and exchange:

- broadcast the model online;

- consult the model online;

- be interoperable with a cadastral plan or other traditional geographic information system (GIS) layers;

- provide a 3D immersion.

– Administrate:

- take decisions on strategic projects;

- re-develop the building to enable visits from the public;

- visualize and present projects during the planning permission applications;

- allow tracking of town-planning regulations.

The conservation survey is essentially an exception to the building specifications. It can be used as a memory in case of deterioration, or even total or partial destruction, which can be due to a range of causes: “war, significant restructuring, but also disrepair and lack of maintenance” [SAI 92]. However, without being able to predict how the survey will be used in this case, it is difficult to establish a comprehensive set of building specifications.

1.2.2. Conversion into building specifications

These potentials should firstly be converted into product specifications to be delivered (nature, coordinate reference system, accuracy of georeferencing, spatial resolution, coverage, lead times, etc). Depending on the formulation of the order, the products to be delivered can be made available in a number of different forms, as shown by the examples below:

– point cloud showing the main structural lines of the building;

– point cloud to enable user to count mason’s marks;

– point cloud with a spatial resolution of 1 cm, with permitted noise of 2 mm;

– 3D line drawing showing the major structural lines of the building;

– 3D line drawing showing the major structural lines of the building and the stone façade;

– triangulated and textured 3D line drawing;

– contour lines giving volume to a dome;

– façade elevations;

– sections, etc.

As well as the aspects to survey, the building specifications must state the precision of the associated geolocalization. The internal precision of the survey must be coherent with the survey’s resolution because, if this is not the case, significant geometric gaps could be interpreted as characteristics of the structure. Demands may often be relaxed on the absolute precision of localization; positioning in a given space is useful in terms of situating the structure in its environment, or to understand the principle axes in relation to the cardinal points for instance. The precision necessary can, therefore, be decimetric or even metric.

If the project sponsor makes the added requirement of communicating, broadcasting and exchanging information on the 3D model produced, then post-processing of the 3D point cloud will be necessary to lighten it, structure it and add semantic information. If the order is more focused on the interoperability of the survey, especially with the requirement of superposing pre-existing data onto the new model, special care needs to be taken regarding the absolute georeferencing. It is, unfortunately, impossible to establish simple links between orders and the product due to the combination of many different parameters that must be taken into account, such as the budget, schedules and preparation, thus making each project unique.

1.2.3. Technical survey specifications

The specifications of the product requested will be converted into technical survey specifications by the contractor: laser and/or image acquisition (resolution, provisional number of scanners or photographic equipment), georeferencing equipment (how, how many and where?), material and software necessary for post-processing [BRY 09].

Evidently, the optimal spatial resolution of images depends on the requirements; to produce a model of a building’s major structural lines, a centimetric or even decimetric resolut...

Table of contents

• Cover
• Contents
• Title Page
• Copyright
• Acknowledgements
• Introduction
• 1 Specific Requirements for the 3D Digitization of Outstanding Sites
• 2 3D Digitization Using Images
• 3 3D Digitization by Laser Scanner
• 4 Complementarity of Techniques
• 5 Point Cloud Processing
• 6 Management and Use of Surveys
• Conclusion
• Index