=Paper=
{{Paper
|id=None
|storemode=property
|title=The Indiana MAS Project: Goals and Preliminary Results
|pdfUrl=https://ceur-ws.org/Vol-892/paper10.pdf
|volume=Vol-892
|dblpUrl=https://dblp.org/rec/conf/woa/LocoroMBMADPTPF12
}}
==The Indiana MAS Project: Goals and Preliminary Results ==
https://ceur-ws.org/Vol-892/paper10.pdf
The Indiana MAS Project:
Goals and Preliminary Results
V. Mascardi, D. Briola, A. Locoro, V. Deufemia, L. Paolino, G. Tortora, R. Francese, G. Polese
M. Martelli, M. Ancona Facoltà di Scienze MM. FF. NN.,
Dipartimento di Informatica, Bioingegneria, Università degli Studi di Salerno
Robotica e Ingegneria dei Sistemi, Email: {deufemia, lpaolino, tortora, francese, gpolese}@unisa.it
Università degli Studi di Genova
Email: {viviana.mascardi, daniela.briola, angela.locoro,
maurizio.martelli, massimo.ancona}@unige.it
Abstract—The Indiana MAS project, funded by the Italian ...
Ministry of Education, University and Research “Futuro in 1) to develop scientific and technical studies and
Ricerca 2010” program, aims at providing a framework for research and to work out such operating meth-
the digital protection and conservation of rock art natural and
cultural heritage sites, by storing, organizing and presenting ods as will make the State capable of coun-
information about them in such a way to encourage scientific teracting the dangers threaten its cultural or
research and to raise the interest and sensibility towards them natural heritage;
from the common people. ...
The project involves two research units, namely Genova
By adhering to some of the most significant objectives
(Dipartimento di Informatica, Bioingegneria, Robotica e Ingeg-
neria dei Sistemi) and Salerno (Dipartimento di Matematica e of the UNESCO World Heritage Convention, the European
Informatica), for a period of 36 months, starting from march Community has recently founded several projects aiming at
8th, 2012. fostering the quality and effectiveness of ICT in the cul-
The technologies adopted in the project range from agents to tural heritage field: the Epoch network of excellence (http:
ontologies, as requested by the complex nature of the platform, //www.epoch-net.org/); the STREP project MultiMatch (http:
where each module is devoted to a specific task: sketch and
symbol recognition, semantic interpretation of complex visual //www.multimatch.eu/) focusing on facilitating the interaction
scenes, multi-language text understanding, storing, classification between users and multi-modal and multilingual online cul-
and indexing of multimedia and heterogeneous digital objects. tural contents; the CIDOC CRM (http://www.cidoc-crm.org/)
All of them should cooperate and coordinate in order to enable and Europeana Data Model (http://pro.europeana.eu/) standard
higher level components to reason on them and to detect efforts, for knowledge representation in cultural heritage;
relationships among different digital objects, hence providing new
hypothesis based on such relationships. the FP6-IST BRICKS project (http://www.brickscommunity.
org/) and the DELOS network of excellence (http://www.
I. I NTRODUCTION delos.info/), both aiming at designing, organizing, integrating
and preserving digital objects in cultural heritage digital li-
According to the UNESCO World Heritage Convention braries; the FP6-IST AGAMEMNON project (http://services.
Concerning the Protection of the World Cultural and Natural txt.it/agamemnon/), where Prof. Massimo Ancona was actively
Heritage, signed in 1972 (http://whc.unesco.org/?cid=175), involved, with the goal of offering to the visitor of an archae-
“cultural heritage” includes: works of man or the combined ological site a personalised and enriched experience through
works of nature and man, and areas including archaeologi- the use of mobile phones; and many other projects.
cal sites which are of outstanding universal value from the Although all such projects represent a source of inspiration
historical, aesthetic, ethnological or anthropological point of for Indiana MAS and share many objectives with it, the
view; natural sites or precisely delineated natural areas of peculiarity of the artifacts handled by Indiana MAS makes
outstanding universal value from the point of view of science, it necessary the adoption of specific tools and technologies
conservation or natural beauty that are defined as “natural that are not applicable in the above projects, due to their more
heritage”. general scopes.
To cite the UNESCO World Heritage Convention The aim of Indiana MAS, funded by the Italian Ministry of
To ensure that effective and active measures are Education, University and Research “Futuro in Ricerca 2010”
taken for the protection, conservation and presenta- program, is the development of a framework for the digital
tion of the cultural and natural heritage situated on preservation of rock art, able to complement the techniques
its territory, each State Party to this Convention shall adopted for the cultural heritage field with the adoption of
endeavour, in so far as possible, and as appropriate context specific techniques.
for each country: Rock carving art is an example of cultural and natural
heritage, since it is often located in wonderful natural sites of such data into an existing collaborative tool set, and it
and represents an invaluable resource for understanding our should supply domain experts with collaborative facilities for
history. It is not a case that the well known rock carving processing the data and making assumptions about the way
sites of Tanum in Sweden, Altamira is Spain, Alta in Norway, of life of the ancient people based on these data. The digital
Lascaux in France, Valcamonica in Italy are all listed in the preservation, classification, and interpretation of rock carvings
UNESCO World Heritage Sites. raises many research challenges, such as the integration of
Many other rock carving sites are spread all over Europe data coming from multiple sources of information and the
and from Northern to Southern Italy (Ciappo delle conche, interpretation of drawings whose meaning may vary based on
Ciappo dei ceci, Ciappo del sale, Pietra delle coppelle, Monte several information such as the objects depicted in the whole
Beigua in Ligury; Val Camonica and Valtellina in Lombardy; carving.
val Chisone, val di Susa, Val Sangone, Val Maira in Piedmont; The solution we propose to suitably face these challenges
Monte Sagro in Tuscany; Lillianes in Aosta Valley; Grotta del is based on the integrated use of intelligent software agents,
Genovese and Grotte dell’Addaura in Sicily). ontologies, natural language processing and sketch recognition
The Indiana MAS project aims at the digital protection and techniques. Multi-agent systems (MASs) represent an optimal
conservation of rock art natural and cultural heritage sites, by solution to manage and organize data from multiple sources
storing, organizing and presenting information about them in and to orchestrate the interaction among the components
such a way to encourage scientific research and to raise the devoted to the interpretation of the carvings.
interest and sensibility towards them from the common people. Ontologies allow to define a common vocabulary that can
The platform that will be designed and implemented during be profitably exploited to organize data associated with rock
the project should support domain experts in the creation carvings, included their semantic annotations, and create se-
of the repository, which may become a reference at Italian mantic relationships between them. Natural Language Process-
and, maybe, European level as a thorough database of rock ing techniques can be used to extract relevant concepts from
carvings, and in the interpretation of rock carvings. It should text and for mining semantic relationships among them, hence
also promote the awareness and the preservation of the cultural supporting the definition and evolution of ontologies devoted
treasure by making cultural information accessible to all on the to describe the domain.
Internet and preserve it for future generations. Sketch recognition techniques can be applied to classify
In this paper we illustrate the objectives and the application the elementary shapes of the carving drawings and associate
domain of the project (Section II) as well as the expected their possible interpretations with them, and can also be
results (Section III), and describe the preliminary results used to obtain an automatic interpretation of a symbol that
obtained in the first months of the project activities (Section users draw on their tablet device. In the end, bidimensional
IV). Section V concludes the paper with some reflections and image recognition techniques can be successfully used to
future work. compare different rock carvings reliefs, or reliefs of the same
carving done by different archaeologists, in order to detect
II. O BJECTIVES AND A PPLICATION D OMAIN
similarities between geographically distant carvings and to
Indiana MAS has five main objectives: assess differences and analogies between techniques used by
• O1. integrating heterogeneous unstructured data (multi- different archaeologists in different ages.
lingual textual documents, pictures, and drawings) related The multi-agent framework (“Indiana Multi-Agent System”
to rock carvings into a single repository; or simply “Indiana MAS”) will be general enough to be used
• O2. normalizing data by recognizing those referring to in any cultural heritage domain where rock carving is a central
the same object, correctly associating them with its digital feature. However, in order to demonstrate the feasibility of our
representation, and removing duplicate data; proposal and to measure its results in a quantifiable way, we
• O3. classifying normalized data according to the “Indiana will apply it to the preservation of the rock art of Mount Bego.
ontology” that will be extracted in a semi-automatic way The choice of Mount Bego testbed was made because,
from the unstructured data, and that will evolve as data thanks to a consolidated collaboration between the University
will; of Genova and the Laboratoire Départemental de Préhistoire
• O4. organizing classified data into a Digital Library and du Lazaret, Nice, France, the University of Genova has partial
making the library accessible thanks to a web-based, access to the ADEVREPAM database containing informa-
multilingual, user-friendly interface; tion about all the rock carving reliefs of that site (45.000
• O5. interpreting data stored in the Digital Library, finding records). Also, the University of Genova owns an inedited
relations among them, and enriching them with the se- and invaluable collection of up to 16.000 drawings and reliefs
mantic information extracted thanks to this interpretation made by Clarence Bicknell between 1898 and 1910, in his
and relation retrieval stage. campaigns on Mount Bego. Bicknell’s legacy also includes
To this end, such a platform should enable the preser- nine notebooks, filled with notes in Victorian English, mostly
vation of all kinds of available data about rock carvings, unpublished. The integration of Bicknell’s legacy with the
such as images, geographical objects, textual descriptions of ADEVREPAM database represents the best possible way
the represented subjects, and the organization and structuring for safeguarding Bicknell’s precious work. This is another
important objective of the project. similar documents (text and images) in 35% cases on
Besides these “organizational” reasons, there are also sci- average.
entific ones motivating the choice of Mount Bego as a testbed • We already developed an ontology [1], based on the
for making the project’s objectives and results more concrete. taxonomy by de Lumley and Echassoux [2], for tagging
Archaeologists and historians look at the area around Mount drawings with concepts taken from that ontology (namely,
Bego as an incredibly valuable source of knowledge, due to the for classifying drawings according to the ontology); that
up to 40,000 figurative petroglyphs and 60,000 non-figurative ontology will be extended in a semi-automatic way in
petroglyphs scattered over a large area at an altitude of 2,000 order to correctly classify multilingual documents and
to 2,700 meters. The historical relevance of the Mount Bego pictures, and not only drawings, and will be used to
petroglyphs is unquestionable, as they date back to the early classify newly inserted documents according to it. As far
Bronze Age, when humans left no written evidences and the as textual documents are concerned, we expect that at
only witnesses of their existence are their tools and, indeed, least 80% documents (in English, Italian and French) will
their “drawings”. Mount Bego rocks are not protected in a safe be correctly classified according to the ontology. As far
place such as a museum and thus they are constantly exposed as drawings are concerned, the correct classification will
to rough weather as well as vandalism of careless or malicious be at least 25%.
visitors. If the latter may not be the main source of damage • A Digital Library will be developed according to the
(the Mount Bego area is hardly accessible in wintertime), the current standard formats and accessibility protocols in
first is definitely a constant threat; 8 months a year many of the order to make a part of the knowledge on Mount Bego
petroglyphs are drown into a thick curtain of snow and rains rock art available to everyone. The Digital Library,
are also frequent in summertime. It comes with no surprise named “Indiana GioNS” - Genoa, Nice, Salerno - will
that many petroglyphs have been (and are still being) eroded be hosted by Genova and will be available, via a web
and some have been totally destroyed. and agent-based interface, starting from the beginning of
Finally, there are historical and geological reasons making the project’s second year.
Mount Bego rock art understanding relevant for Italian rock art • The Knowledge represented in this project will be also
understanding too. In fact, until 1947 Mount Bego belonged analysable and visualisable from a spatio-temporal per-
to Italy, and it shows close relationships with carvings that spective, and tools and services will be provided in order
we can find in Italian sites. A section of the engraved area to deal with these aspects. Reference works that will
of Mount Bego still lies on the Italian side and is included inspire us in this direction are [3], [4], [5].
in the Argentera Park. Also, Mount Bego, Monte Beigua, • The Indiana MAS will integrate agents able to analyze
and Monte Sagro share the same role of relevant sanctuaries and interpret drawings, agents able to reason on pictures,
for the ancient Ligures. Finally, strong relationships exist and agents able to understand natural language (in at least
between Mount Bego and Val Camonica rock art sites too: Italian, English, French). From the interaction among
Val Camonica rock is named “permian sandstone”. It is a agents of these three different kinds, more sophisticated
siliceous fine granulated sandstone, heavily polished by the interpretations of documents and correlations among
glacier during the last glacial era: it looks like and it acts them will emerge. These results will be integrated into
as a real natural blackboard. Only one other valley in the Indiana GioNS as well, thus implementing a dynamically
Alps shows similar condition, although the rock there is called growing repository of knowledge. This outcome is much
“pelite”: it is, not by chance, the Mount Bego. more difficult to measure in a quantifiable way than
the previous ones; independent domain experts will be
III. E XPECTED R ESULTS required to assess the quality of the interpretations and
relationships resulting from this activity.
According to the objectives introduced in the previous
Section, the results expected from the Indiana MAS project IV. P RELIMINARY R ESULTS O BTAINED
are the following:
In the next sections we are going to illustrate the results
• Integration of Bicknell legacy, written documents (in
obtained in the very first months of the project activities, in
English, Italian and French) and pictures with the Ade-
particular since the proponents received the notification of
vrepam database. As a measurable indicator, we assert
funding (September 2011), by following the work package
that we will add to the Adevrepam database data (already
structure reported in the Gantt shown in Figure 1. The work
storing about 55,000 documents relevant for Mount Bego
package relative to the integration of the components in the
rock art) 10,000 new data by the end of the project.
initial and final Indiana MAS prototypes did not start yet.
• Bicknell legacy contains drawings and annotations of
petroglyphs which are already stored in the Adevrepam
A. Project Management
database. The Indiana MAS needs to recognize duplicates
in order to avoid the creation of multiple separate entries The communication among all the parties takes place easily
for the same object. We expect that the Indiana MAS and efficiently via the indianaMAS@unige.it mailing list and
will be able to automatically recognize duplicate or very via Skype.
Fig. 1. Work Packages: Gantt chart.
The work plan is fully respected (many tasks have already
started even if their official start date is month 5), as shown
by the results achieved in the following sections.
As a software application for the project management we
are currently using both DropBox and Google Drive, since we
are not yet developing software in a joint way and a shared
repository is enough. We will install SVN or other software
management applications if needed.
The Kick-off meeting has successfully taken place on May
25th, 2012, in Capri, associated with the “Advanced Visual
Interfaces” conference where the first Indiana MAS paper [1]
was presented.
B. Indiana MAS Design
Figure 2 depicts the Indiana MAS architecture as it has
Fig. 2. High level architecture of the system.
been conceived in the project proposal, together with its
main components (all the agents interact with each other;
communication arrows have been omitted for sake of clarity).
To refine this architecture and to clarify the functionalities stage of requirements analysis we consider having “registered
that will be offered by our system, we have conducted a users” and “simple users”. The former will insert new data
requirements analysis starting from what kind of information into the Indiana GioNS library, while the latter will only
we will store and manage, how these are related and how they query it. The system interface will handle different languages,
will be available to the end users. Instead of listing all the namely Italian, English and French. In Figure 3 the homepage
functionalities using a text representation or a Use Case form, is shown.
we developed a prototype of the Web Interface for the system. The “new digital object insertion functionality” will be
Such interface is available online (http://www.disi.unige.it/ provided to the “registered users” only. Every digital object
person/MascardiV/Download/IndianaMAS/ReqAnalysis/) and inserted into the library will be characterised by a set of
its mere goal is to present the requirements, that is, to define common metadata, for example Title, Object, Description, and
the main Indiana MAS functionalities that we plan to offer to a list of optional common metadata, for example Author, Age,
the users. GPS coordinate; the “insertion functionality” will be different
The system will represent different types of users: at this for each specific digital object, either text or image.
the user is able to specify many kinds of queries, based on all
the metadata described before. Queries will be performed:
• by entering common metadata,
• by entering image specific metadata,
• by entering text specific metadata,
• by inserting an image to be compared with the already
exiting ones,
• by inserting a text to be compared with the already
existing ones,
• by using AgentSketch to draw a sketch to be compared
with the other images.
In addition, all these queries can be combined. Queries can be
executed only over the Indiana GioNS library or over a set of
related libraries, selected by the user from the ones exposed
by the metadata harvester (see the next Section for details).
The “simple users” will be able to follow a procedure called
“analysis”, that is the same procedure of the insertion, but
without inserting a new object into the digital library. This
procedure will help the user to evaluate an image or a text
against what already exists in the system, by taking advantage
Fig. 3. Interface Prototype devoted to the Requirements Analysis.
of the Indiana MAS support in extracting features, without any
desire/permission to insert this object into the library.
For example, text related metadata are relative to the Type C. Design and Development of the Indiana GioNS Digital
of text (Article, book, Master thesis etc.), the Language, the Library
Abstract, and so on. These metadata will be a superset of a During the last months we were actively involved in a
standard language for the description of digital objects such project for the design and development of a Digital Library
as Dublin Core1 . Management System called MANENT [7]. The system is
The image related metadata are relative to the Type of image evolving towards an innovative architecture that is based
(for example, whether it is a colored image or a black and on big data management systems that run upon distributed
white image, whether it depicts something real or it is a manual frameworks such as Hadoop2 and HBase3 . These technologies
drawing, a panorama, and so on), the Symbol(s) appearing in are exploited by social networks such as Facebook, Twit-
the image, their Interpretation. ter, and LinkedIn and everywhere a support for the effec-
All the values allowed for the metadata will be chosen from tive analysis and computation of TeraBytes of information
the Indiana ontology. Furthermore, the system will allow the is needed. Following Google BigTable database4 , the open
users to specify related digital objects already inserted into the source community developed the Hadoop high-scalable dis-
repository. The AgentSketch [6] functionality will be provided tributed filesystem that is de facto a clone of BigTable, and is
to manually draw a picture (that will be then treated as an based on the MapReduce algorithm for the creation of parallel
image): this can be useful to insert a new relief or to trace a tasks that work in each datanode of a cluster, without the
relief using a real picture as background. need for a developer to implement the mechanism of task
The system will accept from the user as many metadata distribution, delivery, and execution. The master node, that is
as s/he knows, then an automatic procedure will be called the namenode, checks and handles automatically the different
to fill and check all such metadata, with the aim to deduce processes that run on each datanode of the system.
more information from the digital object than the user knows, MANENT is the core of a digital library that should contain
helping him to link this new object to those already existing a metadata harvester for collecting information on the digital
in the library, hence to infer new information about the object repositories that all over the world expose their data and
itself. For example, Indiana MAS will analyse the image and metadata in a standard format, such as those provided through
will recognise the atomic symbols it contains, and will suggest the OAI-PMH protocol5 , as well as proper information coming
a possible interpretation of them, while reporting the list of from its local digital library. By following the Digital Library
similar digital objects. Reference Model provided by the DELOS project6 , the system
After this process terminates, the user can either accept the 2 http://hadoop.apache.org/.
metadata automatically extracted by Indiana MAS or modify 3 http://hbase.apache.org
them, before confirming the insertion of the new digital object. 4 http://en.wikipedia.org/wiki/BigTable.
The query phase follows the above mentioned data structure: 5 http://www.openarchives.org/OAI/openarchivesprotocol.html.
6 http://www.delos.info/index.php?option=com content&task=view&id=
1 http://dublincore.org/. 345.
Fig. 4. the MANENT Architecture.
should provide functionalities for the collection, organisation, 2) Training the SOM by using the extracted features;
search, and browsing of digital resources, such as documents, 3) Classifying the query drawing Q by looking for the map
images, people, and external sources of knowledge. cluster in the SOM most similar to the features extracted
The Indiana GioNS Digital Library should be hence from Q.
equipped with semantic services for the multi-language text In order to verify the effectiveness of such descriptors, we
analysis, automatic classification, similarity matching, seman- evaluated the constructed SOM on three datasets: MPEG7,
tic annotation, and indexing of all the sources of information Relief75 and Relief1400. MPEG7 shape database is one of the
collected inside the library and all over the world. most popular dataset used for comparing the performances of
Figure 4 depicts the high-level architecture of the MANENT image similarity algorithms [12], and it contains about 1400
system that should be the core of the digital library manage- images divided into 70 classes. Relief75 and Relief1400 are
ment system developed inside Indiana MAS. The core com- two datasets built by ourselves to understand the effectiveness
ponent for the local digital library is organised according to of our approach on the project test-bed, namely the Mont Bego
different standard languages for the description of repositories reliefs. The former contains 75 images divided into 10 classes
and digital objects (such as Dublin Core, EAD7 and Marc8 ). In while the latter contains 1400 images derived by the previous
addition, it provides functionalities to the user for the insertion 75 by means of a distortion algorithm.
of new objects and the search and browsing of them according The results we collected at the end of the experiments are
to such descriptions. Different adapters are finally conceived shown in Figures 5, 6, and 7:
to interface the system with the below cluster infrastructure
and the above web interface.
D. Design and Development of Image Analysis, Classification
and Interpretation Agents
In order to analyse and properly classify the relief drawings
several similarity measures have been evaluated so far. In
particular, we analysed three descriptors which have been
widely used for shape recognition: Shape Context [8], Inner-
Distance Shape Context [9], and Radon transform [10]. To
allow a fast and effective drawing classification, the use of
such descriptors has been combined with Self-Organizing Map
(SOM) [11], a clustering and data visualization technique
based on neural networks. In general, the process used for
relief classification can be organized in three steps:
Fig. 5. Percentage of MPEG7 correct classifications considering: the top
1) Extracting the features from the drawing dataset based scored class, the top-2 scored classes, and so on.
on the considered descriptor;
7 http://www.loc.gov/ead/. It is worth to note that no algorithm generally provides
8 http://www.loc.gov/marc/. optimal classification in the first hit, but a good classification
language. Also, it can be run as a service in the cloud. MUSE,
whose design is completed and whose implementation is under
way, will be soon experimented in the Registry Office of
Genoa Municipality. A sketch of the different components of
the system is reported in Figure 8
Fig. 6. Percentage of relief75 correct classifications considering: the top
scored class, the top-2 scored classes, and so on.
Fig. 8. The MUSE system.
Some modules of the MUSE system have been integrated
as black box items (e.g. Google Translate). Once a translation
is made from the user query formulated in his native language
to the Italian language, a query expansion procedure is run to
help disambiguate the request and find a match with one of the
“well-known problems” encoded in the MUSE ontology. The
aim of the ontology is that of driving the system to retrieve the
correct procedural rule against the user query, once the query
has been properly expanded and interpreted. To this aim, the
ontology contains all the different paths elicited by the domain
experts. If users do mistakes when formulating their queries
Fig. 7. Percentage of relief1400 correct classifications considering: the top
scored class, the top-2 scored classes, and so on.
MUSE records such mistakes and the relative correction, and
ranks the pairs obtained as they are repeated during real-time
interactions. In this way the system increases the strength of
its hypothesis on the occurrence of such patterns more often
can be obtained in a semi-automatic way considering the top- than chance, and may provide an automatic correction when
scored results of the search. More precisely, by considering the the same situation happens again, as well as top ranking the
three descriptors and the three datasets, the Radon provides the more probable translations associated with the more probable
correct class into the first 4 hits in almost the 100% of cases requests.
while, a less precise solution can be achieved by considering
only 3 hits. In this case, the correct class appears into the first F. Dissemination of Results
three hits in the 80% of cases. The dissemination activity has been conducted by creating
the project web site http://indianamas.disi.unige.it, and by
E. Design and Development of the Agents devoted to the
publishing the papers cited in the above paragraphs.
Analysis, Classification and Interpretation of Multilingual
The Indiana MAS project has been mentioned in the seminar
Documents
with title “Le incisioni rupestri al Monte Bego nei rilievi di
Some work in the direction of multilingual services was Clarence Bicknell 1906-1917” held on April 26th, 2012 in
accomplished during the design and development of the MUSE Genova by Antonella Traverso (“Soprintendenza per i Beni
project [13] that addresses the problem of the provision of Archeologici della Liguria”) and Cristina Bonci (Università
multilingual services in the domain of Public Administra- degli Studi di Genova) at the “Accademia Ligure di Scienze e
tions, supporting business and interpersonal communication Lettere”.
and enabling people to make sense of content and services A PhD course on “Computational Archaeology” is foreseen
already available in this domain. MUSE exploits state of the art in 2013, based on the expertise acquired during the project.
machine translators, a formalized domain description ontology,
a flexible and distributed architecture based on intelligent V. C ONCLUSIONS AND F UTURE W ORK
agents, a set of stepwise procedures codified in form of The project has just started (the official date is march 8th,
plans in an existing declarative agent oriented programming 2012), but the results already obtained are meaningful and
encouraging. [4] G. Mantegari, M. Palmonari, and G. Vizzari, “Rapid prototyping a
The project goals are very ambitious, due to the combination semantic web application for cultural heritage: The case of mantic,” in
The Semantic Web: Research and Applications, ser. LNCS, vol. 6089.
of multi-modality, multi-linguality, and heterogeneity of the Springer, 2010, pp. 406–410.
data, and the necessity to give a semantic interpretation of [5] A. Bonomi, A. Mosca, M. Palmonari, and G. Vizzari, “Integrating
them based on sophisticated reasoning. It was hence decided a wiki in an ontology driven web site: Approach, architecture and
application in the archaeological domain,” in Proceedings of the
to face some work before the official starting date, in order 3rd Semantic Wiki Workshop (SemWiki 2008) at the 5th European
to have time to widely experiment all the single elements of Semantic Web Conference (ESWC 2008), ser. CEUR Workshop
the system, and to design and develop all the rules necessary Proceedings, vol. 360. CEUR-WS.org, 2008. [Online]. Available:
http://ceur-ws.org/Vol-360/paper-13.pdf
to infer new knowledge starting from the objects stored and [6] V. Mascardi, V. Deufemia, D. Malafronte, A. Ricciarelli, N. Bianchi, and
analysed inside the system. Such rules are so complex that H. D. Lumey, “Rock art interpretation within Indiana MAS,” in Proc.
a strict and long-term collaboration with domain expert is of KES 2012, 2012.
[7] A. Locoro, D. Grignani, and V. Mascardi, MANENT: An Infrastructure
unavoidable. for Integrating, Structuring and Searching Digital Libraries, ser. Studies
Future works are those envisioned for each work package in Computational Intelligence. Springer, 2011, vol. 375, pp. 315–341.
during the proposal step. At the moment no deviation from the [8] S. Belongie, J. Malik, and J. Puzicha, “Shape matching and object
recognition using shape contexts,” IEEE Trans. Pattern Anal. Mach.
work plan seems necessary. The only limitation has emerged Intell., vol. 24, no. 4, pp. 509–522, 2002.
during the experiments conducted by Alessandro Ricciarelli in [9] H. Ling and D. Jacobs, “Shape classification using the inner-distance,”
his Bachelor’s Thesis, and described in [6]. Such experiments IEEE Trans. Pattern Anal. Mach. Intell, vol. 29, pp. 286–299, 2007.
[10] B. Rubin, “Radon, cosine and sine transforms on real hyperbolic space,”
were further conducted by the Salerno Unit, and are focusing Advances in Mathematics, vol. 170, no. 2, pp. 206–223, 2002.
on the possibility of recognising and classifying pictures of [11] T. Kohonen, Ed., Self-organizing maps. Springer, 1997.
engravings. The rock veins, the presence of stains and moss, [12] S. Jeannin and M. Bober, “Description of core experiments for mpeg-7
motion/shape,” Tech. Rep., 1999.
and the light effect on the irregular surface of the rocks make [13] M. Bozzano, D. Briola, D. Leone, A. Locoro, L. Marasso, and V. Mas-
quite impossible to distinguish the engraving trace from noise. cardi, “Muse: Multilinguality and semantics for the citizens of the
This challenge has been also highlighted in [14] where the world,” in IDC 2012, submitted, 2012.
[14] Q. Zhu, X. Wang, E. Keogh, and S. Lee, “Augmenting the generalized
authors state that the extraction of petroglyph contours within hough transform to enable the mining of petroglyphs,” in Proc. of KDD
pictures is a very challenge task, even in the case of images ’09, 2009.
with a high and clear contrast. As a consequence, automatic
recognition and classification of petroglyph pictures where
contours are difficult to delineate even by people is almost
impossible to perform. For this reason we will concentrate on
black and white relief images and will postpone the treatment
of pictures to a future activity.
ACKNOWLEDGEMENTS
The FIRB project “Indiana MAS and the Digital Preser-
vation of Rock Carvings: A multi-agent system for drawing
and natural language understanding aimed at preserving rock
carvings” is funded by the Italian Ministry of Education,
University and Research under fund identifier RBFR10PEIT.
The authors would like to thank Daniele Grignani for the
support in the design and development of the MANENT sys-
tem, Prof. Henry de Lumley for sharing digital material useful
for the project and Nicoletta Bianchi and Antonella Traverso
for their precious availability in providing information and
feedback on the real needs of the archaeologists.
R EFERENCES
[1] V. Deufemia, L. Paolino, G. Tortora, V. Mascardi, M. Ancona,
M. Martelli, A. Traverso, N. Bianchi, and H. de Lumley, “Investigative
analysis across documents and drawings: Visual analytics for archeolo-
gists,” in Procs. of AVI 2012, 2012.
[2] H. de Lumley and A. Echassoux, “The rock carvings of the chalcolithic
and ancient bronze age from the mont bego area. the cosmogonic myths
of the early metallurgic settlers in the southern alps,” L’Anthropologie,
vol. 113.5P2, pp. 969–1004, 2009.
[3] T. Kauppinen, G. Mantegari, P. Paakkarinen, H. Kuittinen, E. Hyvönen,
and S. Bandini, “Determining relevance of imprecise temporal intervals
for cultural heritage information retrieval,” International Journal
of Human-Computer Studies, vol. 68, no. 9, pp. 549 – 560,
2010. [Online]. Available: http://www.sciencedirect.com/science/article/
pii/S107158191000039X