Visual Analysis of Architectural Heritage:

The Interior Décor of the Domus of Roman Tunisia

Aida Hermi-Nasr

, Najla Allani

and Jean-Yves Blaise

National School of Architecture and Urbanism, University of Carthage, Tunis,Tunisia

UMR CNRS/MCC 694 MAP, Marseille, France

Keywords: Domus of the Roman Tunisia, Inventory, Comparative Analysis, Architectural Modelling, Information

Visualization.

Abstract: This paper aims to propose a new approach which can bring a renewal of the means of study of the Domus of

Roman Tunisia. We had chosen thirty roman houses built from 146 (B.C) to 439. These houses are spread

over 19 Tunisian cities. The paper is based on an approach called information modelling, which is situated at

the interface of the architectural modelling and the Information. This study is built on a numerical

implementation, which allows us to test some methods of analysis based on the group of the gathered cases.

It proceeds, following a method inventory prior to these houses to a comparative analysis by focusing on the

mechanisms of visual comparison between the Domus. In one hand, the study tries to structure and save large

volumes of data which are generally heterogeneous, doubtful, incomplete and sometimes contradictory. In

the other hand, the study attempts to preserve the history of the architectural evolutions. Applying this method

of work on the suggested cases permits to focus on the regularities and the individual and collective evolutions,

to emphasize the convergences and the divergences between the edifices and the periods. Finally, it permits

to improve the exchange of knowledge between experts.

1 INTRODUCTION

In recent years, under the influence of development

of NICT, the study of the architectural heritage is a

sector in full revaluation. This study of architectural

heritage has undergone profound changes, with the

introduction of new information management tools

and with the application of new technologies of data

acquisition and volumetric representation of the

building. This research aims to characterize Roman

architecture through a recognition, identification and

modelling procedure. It concerns the Roman

domestic architecture of Tunisia from 146 (B.C) to

439 as an experimental ground (this point is discussed

later). We assume that this type of architecture lends

itself well to a new scientific study (Bullo et al.,

2013).

This research proposes a new conception of

heritage information systems based on tools in the

areas of database management. Digital modelling is

as an essential tool for advancing knowledge. Works

are developing primarily in two directions: The

design of heritage information systems leveraging

tools in the areas of database management systems.

The inventory makes it possible to set up a recording

and data management tool (Morand, 2002). The

second part deals with comparative analysis and

information visualization through which we will try

to answer, through practice (s), some of the issues

raised by the introduction of new technologies in the

study of heritage buildings.

The method we introduce identifies three key

elements:

 A formal analysis of the architectural typology in

order to define the parameters we need to

compare, and the visual sign in charge of

conveying comparisons. Result of this first step is

a univocal graphic symbolization of the

typology’s architectural composition and a

relational database containing the data set itself

(Hermi et al., 2014).

 A database of Domus of Roman Tunisia, in order

to investigate an architectural inventory approach,

multidisciplinary, authorizing an operational

chaining between tools, approaches, and

disciplines dispersed (Hermi et al., 2014).

 A set of visual signs, calculated dynamically for

each object in the data.

Generally, we try to explain, using a data set

concerning antique houses, that visual comparative of

Hermi-Nasr, A., Allani, N. and Blaise, J.

Visual Analysis of Architectural Heritage: The Interior Décor of the Domus of Roman Tunisia.

DOI: 10.5220/0008261004050408

In Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2019), pages 405-408

ISBN: 978-989-758-382-7

405

the data can provide a major methodological

innovation for questioning of information on

architectural objects. This research sought to illustrate

how infovis methods can be fruitfully applied in the

architectural heritage. Throughout this article, we try

to present three important steps: Morphological

analysis, Elaboration of a Database and visual tools.

2 PRESENTATION OF DATA SET

AND METHOD

Roman domestic architecture presents a heritage from

an architectural know-how that we must try to

analyse, recognize and understand its evolution. The

data set we have chosen is an ensemble of 32 Domus

of Roman Tunisia and in time from 146 (B.C) to 439.

The first criterion for selecting the material is its state

of conservation which must be adequate for this type

of analysis. Roman houses, a typical element of

antique towns, have common features wherever they

are, and whenever they were built: their architectural

composition (three major structures: Open spaces is

the atrium was the open central court with enclosed

rooms on all sides, A triclinium is a formal dining

room in a Roman building, and residential rooms).

But elements of variation in the Domus typological

family are numerous. Still it is important to stress that

if we are to enable visual comparisons (and thereby

maybe better understand the variations), we need to

clearly delineate for each structure its specific

parameters. Naturally, parameters exist, such as date

of construction, which qualify the Domus as a whole.

The specificity of our approach appears to us rather

in the fact that inside the Domus we identify

significant architectural structures that we analysis

formally in step 1 (Morphological analysis) of the

proposed method. In step 2 (Elaboration of a

Database) we focus on developing a heritage

inventory of corpus spread over the Tunisian

territory. Finally, we try to develop a set of visual

tools, it allows us to gaining insight on each object

and on the data set as a whole by fostering one-to-one

or several-to-several comparison and characterising

each object by providing a univocal visual signature

not of the object itself but of the information behind

the object (Blaise et al., 2006).

2.1 Morphological Analysis

Morphological analysis of the architectural typology

can be described as an investigative process in which

we first identify main structures. This step allows

understanding and formalization of objects of

knowledge in order to show their characteristics. This

method seeks to determine the morphological identity

of the studied object by constructing one or more

structural models of the object under consideration.

Consequently, we have performed a decomposition of

the architectural object into segments according to the

manifest discontinuities and then, we have tried to

regroup the Domus having the same structures of

segmentation. This stage will be followed by a

statistical analysis (degree of absence or presence of

segments). The analysis that we conducted allowed us

to identify the structural models relating to the

disposition of segments and the nature of the shapes

of components, see Fig. 1. We have schematized this

model according to a simplified representation. This

led to an organized model. We obtained from the

analysis three distinct families, see Fig. 2.

Figure 1: The process of structural analysis.

Figure 2: Structural model for each major category.

2.2 Elaboration of a Database

The functional and technical analysis, allowed to list

all available data. Now they must be organized in a

model data. Merise method provides a framework to

enable this analysis and optimize the creation of the

model. The analysis method Merise (Baptiste, 2009)

was created in the late 70s, by the will of public

authorities wishing to provide the government and

public enterprises with a rigorous methodology. It

incorporates new features: computer database. There

For Domus: Denomination, Floor underground

Upstairs, Total surface area of the Domus, urban type,

maximum length Domus, maximum width Domus,

Orientation Domus, Delimitation Position Domus

and general presentation. The steps of the elaboration

of the database are:

1. Designing a conceptual data model without

considering the computational aspects;

2. The implementation of a logical data model

depending on the structure of the computer

software selected;

KDIR 2019 - 11th International Conference on Knowledge Discovery and Information Retrieval

406

3. And the establishment of a physical model of

data with the language used in the software.

Figure 3: General structure of the method of constructing a

database.

The data in the database are managed in a designed

system that imposed its performance in managing

databases relationally model. This tool is the SGBDR

MySQL (Server of relational databases, software

widely used for the manipulation of textual data), that

is part of management software database most used in

the world. This tool provides a work environment

because of the wide variety of elements based on

language SQL that he provides. While the interface to

manage data MYSQL performed on a server

PHPMyAdmin constitutes an excellent tool for server

administration MySQL (Meier A., 2006). This handy

interface allows you to run, easily and without much

knowledge in the area of databases, many requests

such as creation of data table, inserts, updates,

deletes, changes in the structure of the database. This

system is very useful for backing up a database as

SQL file and easily transfer their data.

After the analysis phase, it is possible to develop

a conceptual data model that integrates the different

reference tables and which taking into account the

needs of users leaving a possible subsequent changes

if necessary. The database is structured around

entities. These entities are used to organize in a

relevant and practical way the information on the

Domus of Roman Tunisia. Each entity must have a

code corresponds to a unique identifier for each

object.

Take the case of this two entities, which have the

following properties:

 For Domus: Denomination, Floor underground

Upstairs, Total surface area of the Domus, urban

type, maximum length Domus, maximum width

Domus, Orientation Domus, Delimitation

Position Domus, general presentation;

 For sites: Current name, Latin name, area of

town, area of the city, Year of inscription by

UNESCO, Registered by UNESCO, current

urban Configuration, Punic, and Azimuth

Domus.

This part is used to establish a database of Domus of

Roman Tunisia from 146 BC until 439 before from

heterogeneous data need rigor and planning during

the analysis phase.

3 VISUAL TOOLS

Roman Domus is a construction rich in structure as in

decoration. In this research, we apply our analysis

protocol to the décor. Up to here what we have done

is distribute in space a visual sign that connects each

to a given implementation of the antique house

typological family. We need to clearly define for each

structure its specific parameters. For this, several

criteria can be taken into account. It is a question of

codifying the material expressions of soil, walls and

threshold. The last step is then to use the parameters

defined in order to permit comparisons. Finaly,

showing through the visual signature of a Domus, the

combination of values for parameters of a given

Domus (Bertin J., 1998), see Fig. 4. This technique

proposes to codify the chosen parameters of the

Domus. For example; Visual site codification, see

Fig. 5, 6.

Figure 4: Visual site codification.

Figure 5: (a) Typical example: Visual graph of the Domus

Protomes, in Thuburbo Minus; (b) Value of the same

parameters across the data set, example: 1-date of

construction, 2-urbain position, 3-orientation, 4-area of

Domus, 5- type of transformation, 6- general type, 7-

situation,8- floor covering,9- threshold coating,10- wall

cladding.

The implementation of the method is based on a

combination of standards or robust technology, and

can be considered as quite straightforward. Graphics

are SVG inside which interactions are simple

Visual Analysis of Architectural Heritage: The Interior Décor of the Domus of Roman Tunisia

407

Javascript commands. The SVG files and their

Javascript associated scripts are produced at query

time by a set of Perl script that read the MySQL inside

which the data set is maintained.

Figure 6: (a) Visual graph of the Domus Triomphe Neptune

in Acholla; (b): A comparison of six Domus having the

same structure (identical morphological configuration,

class II).

We notice that the decoration of each Domus could

fit into a particular type of structure and have a

specific plastic modality. The majority of Domus,

built between the end of the 2nd and the beginning of

the 3rd century, use black and white mosaic and the

mosaic in polychrome tesserae with geometric

designs. During this period the Romans use

architectural and plastic coatings simplified. Houses

that settle in outlying neighborhoods, built between

the 3rd century AD and the 4th century AD, use

architectural coatings with style and patterns floral,

vegetal and figurative. We remark that each

neighborhood has specific stylistic features that could

explain its location.

4 CONCLUSION AND

PERSPECTIVES

In this paper we try to present, using a data set

concerning antique houses, that visual comparative of

the data can provide a major methodological

innovation for questioning of information on

architectural objects. This research sought to illustrate

how infovis methods can be fruitfully applied in the

architectural heritage. Our work is based on three

important steps: Morphological analysis, Elaboration

of a Database and visual tools.

Two benefits of the method we introduce can already

be accentuated:

 Finalize the development of visual codes in

development of other research. The graphics

does not only enable comparison but also

underlines lacks of information. Showing us

unambiguously what we know and what we

don’t know.

 Rectify the organization of the database

More generally, this work, although a lot more needs

to be done, does already underline the gain of insight

researchers in the field of the architectural heritage,

can expect from investing in graphics that say

something. There is no doubt our contribution

remains very isolated show it is worth trying to

exploit architectural analysis using concepts

stemming from infovis. For the future work, we aim

to introduce a three-dimensional visualization,

through which the 3D model plays the role of

interface. Otherwise, the proposed 3D models will

serve as a rich visual interface. They always to

interrogate the base from the scene to quickly build a

scene corresponding to the response to a query on the

database.

REFERENCES

Bullo, S., and Ghedini, F., 2003. Amplimissimae Atque

ornatissimae Domus, résidences des cités de la Tunisie

Romaine. Tome I - Tome II, éditions Quazar, Rome.

Morand, B., 2002. Cours d'analyse et conception des

systèmes d'information. Université de Caen.

Hermi, A., Allani, N., 2014. Strucrural modeling of Domus

of roman Tunisia (146_439 B.C), IWCEA

(International Workshop on Civil Engineering and

Architecture), Istanbul, Turkey, 2014.

Hermi, A., Allani, N., and Blaise J.-Y. 2014. Elaboration

of a database of heritage inventory, JETAE

(International journal of Emerging Technology and

Advanced Engineering) Volume 4, Issue 9.

Blaise, J.-Y, and Dudek I., 2006. Modélisation

informationnelle: un cadre méthodologique pour

représenter des connaissances évolutives spatialisables

, Proc.EGC Conf. (Knowledge Extraction and

Management) Lille, RNTI-E-6, pp.347-358.

Baptiste J-L., 2009. Merise guide pratique, Modélisation de

données et des traitements, Langage SQL.

Meier A., 2006. Introduction pratique aux bases de

données relationnelles, Collection IRIS, Deuxième

édition.

Bertin J., 1998. Sémiologie Graphique, Editions de

l’EHESS, Paris.

KDIR 2019 - 11th International Conference on Knowledge Discovery and Information Retrieval

408