An Automatic Approach for documentation and recovery of rupestrian
paintings using Multisperspectral Remote Sensing
VICENTE BAYARRI-CAYÓN1 , ELENA CASTILLO-LÓPEZ2, JOSE ANTONIO
DOMINGUEZ3
1 GIM GEOMATICS, Torrelavega, Cantabria, SPAIN
2 Department of Geographic Engineering and Techniques of Graphical Expression, University of
Cantabria, 39005, SPAIN
3Department of Applied Geoinformatics and Spatial Planning, Faculty of Environmental Sciences
Czech University of Life Sciences, Kamycka 1176, 165 21 Prague 6 – Suchdol, CZECH REPUBLIC
Abstract: - Today the majority of the research on rock art projects are based on the material remains in the form
of painting and engraving. What exists now are not but the remains or part of that which the hand of man drew
or recorded in the past. The combined action of natural and anthropogenic phenomena have been generating
processes of deterioration which have led to the loss, on many occasions, an important part of the
representations. Preserved signs are that make up the visible remains visually and therefore are the object of
study by specialists. The implementation of a new mathematical techniques that work from data obtained using
current of image acquisition, and mature technologies allow the recovery of representations, to a state of "latent
image" after their deterioration process remain on the surface glyphs support, but are in no way visible to the
human eye, and is the tool base used in the last years to document and analyse these figures. This makes that a
large part of the knowledge generated in this period is only based on a part of the set of representations, the
visible.The current conceptual, scientific and technological capabilities provide us tools to retrieve non-visible
representations, helping to set up new models that tend to contain all of the information shown, changing in a
very significant way representations or models for the study and the dissemination of cave images sets, forcing
us logically
Key-Words: - Rock art, New Technologies, Mathematical Algorithm, Remote Sensing, Hyperspectral Imagery,
Robust Statistic, VNIR.
1.Introduction
The biggest problem on rock art research has
been derived from the location of the
iconographic representations, located in
complex places, narrow sites, caves, or walls of
the rock shelters; also the way in which these
could be represented for its documentation.
This distribution of locations never allowed
easy access and much less provided a
comfortable and thorough study.
The documentation and recovery of rupestrian
paintings has been submitting to the painting
and printmaking techniques of representation
that science and art have been developing at
every moment of history, although almost
never have attempted to rationalize and
standardize the method applied, as well as
assess the ability of representation and distance
with the reality from which it c omes the
theoretical model.
Scientific research needs of specific methods
and therefore all of these systems must be
subject to a methodological consistency, to a
capacity of qualitative and quantitative
evaluation of the represented model and, if
possible, be independent of the subjectivity of
the operator of the system of representation.
For other side, the development of systems and
modelling techniques, mapping and image
acquisition, have achieved that the
representation of the rock art also becomes
mathematical model of the represented object.
The main aim of this article consists of
deploying new methodologies of
documentation of art rock that exceeds the
limitations of the technologies currently
PROOF
DOI: 10.37394/232020.2022.2.5
Vicente Bayarri-Cayón,
Elena Castillo-López, Jose Antonio Dominguez
E-ISSN: 2732-9941
21
Volume 2, 2022
applied to this field of cultural heritage
management.
Fig. 1: Example of training rock art scenery.
2.Methodology
2.1. Objectives of this work
The objectives of this project in relation with
documentation and recovery on rock art are:
a) Determination of the deposits and
decorated panels susceptible of
analysis.
b) Establishment of measurement
practices, depending on t he
characteristics of the furnished study
panels, will establish the corresponding
measurement protocol, ensuring the
proper conservation of rock art.
c) Data collection campaign which include
field work and raw data computer
processing.
d) Development of new technologies. New
mathematical approach using
multisperspectral remote sensing.
e) Analysis and evaluation of results,
approval of the methodology
experimental developed or, if not, the
failure detection of the applied
procedure and implementation of the
adjustments or improvements needed.
2.1. Radiometric measurements
The methodology used to reach our aim has the
following parts: the first one would be to know
the spectral answer of different land bodies.
First of all, we have to know the spectral
response of some representative bases. The
radiometric measures were made with an
analytical spectral devices–full resolution
(ASD-FR) spectroradiometer, equipped with
optical fiber cables.
The ASD-FR was provided by the Centre of
Studies and Research of Public Works
(CEDEX), belonging to the Spanish Ministry
of Public Works and Economy.
The main variable, remote sensing reflectance,
was obtained following NASA’s protocols
(Fargion et al., 2000) and applying the
corrections for specular reflections (Mobley,
1999).
Fig. 2: ASD-FR Spectroradiometer working in
field.
2.2. Multispectral data acquisition
The capture of multispectral information has
been using a prototype camera developed by
the GIM Geomatics company in collaboration
with a german company.
Fig. 3: IRCAM-GIM Sensor.
Different wavelengths were recorded in field
campaigns. The first configuration was aimed
at the capture of data in the infrared, where
appear mixed reflection and emission
processes in variable percentage in function of
the part of the spectrum where it is observed.
For the observation was employed a objective
PROOF
DOI: 10.37394/232020.2022.2.5
Vicente Bayarri-Cayón,
Elena Castillo-López, Jose Antonio Dominguez
E-ISSN: 2732-9941
22
Volume 2, 2022
of an alloy of MgO of 50 mm of focal length
which presents a window observation between
2000 and 4500 nm. The wavelengths used were
2200 nm, 2500 nm , 3050 nm , 3800 nm and
4600 nm.
Also study panels have been captured using a
digital camera Sony Alpha 700.
2.3. Geometric Characterization of the cave.
The geometry of the training zones
characterization has been using a laser scanner
photon of FARO which is capable of
topographically recording the cave with a
precision of 3 mm.
Fig. 4: Scanning made with laser scanning.
2.4. Mathematical Algorithms
The algorithms of digital image processing
applied for this study are available in the
package LIBRA.
LIBRA is a MATLAB Library for Robust
Analysis which is developed at
ROBUST@Leuven, the research group on
robust statistics at the KU Leuven.
It contains user-friendly implementations of
several robust procedures. These methods are
resistant to outliers in the data. Currently, the
library contains functions for univariate
location, scale and skewness, multivariate
location and covariance estimation (MCD),
regression (LTS, MCD-regression), Principal
Component Analysis (RAPCA, ROBPCA),
Principal Component Regression (RPCR),
Partial Least Squares Regression (RSIMPLS),
classification (RDA, RSIMCA), clustering,
outlier detection for skewed data (including the
bagplot based on ha lfspace depth), and
censored depth quantiles.
We have also tested other algorithms
implemented by the research department of
GIM Geomatics.
All these algorithms are implemented in the R
language (R: A Language and Environment for
Statistical Computing).
3.Results
3.1. Methodology
The method of observation and analysis has
been the integration of three technologies, the
information on c olor captured by a digital
camera Sony Alpha 700, t he prototype of
IRCAM-GIM multispectral sensor, with a
different configuration steps of band and
optical as well as laser scanner 3D Faro
Photon, with a length of wave is 785 nm.
Fig. 5: Methodology for capturing information.
We captured information in the visible,
infrared spectrum near and medium, using for
different instrumentation in order to be able to
create a s et of data that analyze altogether.
Spectral responses were recorded to identify
differences in pigments and to determine the
amount of different employees. Other analyses
were treated determine excess paint, you
touch-ups, break when recording,
superimposition of paintings, extraction of
boundaries blurred or covered with calcite,
characterization of bases and paintings on
them.
A number of methodologies has been
developed, tested, inspected and validated
using the recommendations of the
archaeological team.
These methodologies allow to analyze the
signal of the spectral information captured, for
be able to draw conclusions about them.
PROOF
DOI: 10.37394/232020.2022.2.5
Vicente Bayarri-Cayón,
Elena Castillo-López, Jose Antonio Dominguez
E-ISSN: 2732-9941
23
Volume 2, 2022
Fig. 6: Workflow of this study.
3.2.- Recognition of colouring matter
Traditionally the study of the composition of
colouring matter needed to be taking samples,
so this involves catching action. The reading of
the images worked to different spectral
amplitudes can differentiate clearly
compositions different from the coloring
materials used. It is an important milestone.
Fig. 7: Recovery of pigments.
Documented spectrally different zones, in the 5
image may be related clearly with colouring
matter of different mineralogical composition.
The development and deepening of this area
will allow, in the medium to long term, gain
knowledge, at least relative, of coloring
without composition to conduct sampling.
3.3.- Recognition of superposition of forms
Art studies Rock, and conditioned by the State
of conservation of the figures, the reading of
overlapping strokes or figures is one of the
more pronounced problems.
Fig. 8: Visible adquisition.
Fig. 9: Superposition of forms using automatic
approach.
The employee system has enabled the overlays
in at least three (fields: to) between strokes of a
same figure in those cases in which the
composition of colouring matter is different;
(b) between engraving and painting, since they
discriminate spectral and perfectly each one of
the technical actions; (and, to a lesser degree,
c) the overlap between recorded strokes of the
same figure, allowing the reconstruction with
some reliability in the process of execution of a
motive, although this is an end which must be
deeper.
PROOF
DOI: 10.37394/232020.2022.2.5
Vicente Bayarri-Cayón,
Elena Castillo-López, Jose Antonio Dominguez
E-ISSN: 2732-9941
24
Volume 2, 2022
3.4.- Reconstruction of the cave grounds
Most of the figures located in the interior of
cavities are subject to tafonómicos to cause
loss of colouring matter or the erosion of
etched surfaces to they involve a reading, in
many cases, little defined reason.
Fig. 10: Area to rebuild.
Fig. 11: Reconstruction of the art rock.
The application of the technique has allowed to
define accurately the original morphology of
some figures, both recorded as painted in
different colors (and most likely reasons
different chemical composition).
Fig. 12: Discrimination of the pigments and
reconstruction of the motive.
In particular, it is possible to define with
accuracy the outlines of the figures, recognize
with precision anatomical parts or areas of
figures concrete and, accordingly, to obtain
images posed by reconstruction highly reliable
painting or the original print.
In addition, the technical application proposal
has enabled a v ery precise reading (for the
differentiation with the support) of the figures
recorded, to the point of defining with full
accuracy discrimination among the engraved
grooves and fissures, cracks, etc. of the
support.
Thus, figures that in currently presented
difficulties of display can be "rebuilt" and this
mode allow precise formal studies or even
serve as efficient support for the realization of
facsimiles.
PROOF
DOI: 10.37394/232020.2022.2.5
Vicente Bayarri-Cayón,
Elena Castillo-López, Jose Antonio Dominguez
E-ISSN: 2732-9941
25
Volume 2, 2022
4. Conclusions
The application of the data capture system to
the scope of the Archaeology, and in particular
of rock art, presents a high-potential
applications in documentation, technical
analysis and implementation process and in
conservation. In addition and as a result, is an
important tool for the reconstruction of
"images" of the rock art of the moment in
which the figures were laid down so it It
involves in the enhancement of this heritage.
5. Acknowledgements
This paper has been made thanks to
Government of Cantabria.
The acronym of the project is IDICAN
“Estudio de la aplicabilidad de un pr ototipo de
cámara multispectral en el ámbito medioambiental
y arqueológico” (Convocatoria IDICAN).
References
[1].AUJOULAT N.; GENESTE J.M.; RIGAUD
J.PH. Y ROUSSOT A. (1991): La vézère des
origines, paris, éditions du patrimoine, 144 p.
(collection «Guides Archéologiques de la France»)
[2].AUJOULAT, N. (2004): Lascaux, le geste,
l’espace et le temps, Paris, le seuil, 273 p.
(Collection « Art Rupestre »)
[3].BARRIL VICENTE, M.; HERAS MARTIN, C.
(1990): La conservación de la cueva de Altamira:
pasado y futuro, en Journées Internationales
d=etude sur la Conservation de l`Art Rupestre
(périgord, 2023 août 1990): 129-138, périgueux,
icom
[4].BRUNET, J. Y VOUVÉ, J. (1996): La
conservation des grottes ornées, Paris, ministère de
la culture; cnrs éditions, (Collection « Conservation
du Patrimoine »)
[5].C.; SÁNCHEZ-MORAL, S.; SOLER, V.
(2003): Origen bacteriano de espeleotemas tipo
moonmilk en ambiente kárstico (Cueva de
Altamira, Cantabria, España). Estudios geológicos,
nº 59, p. 145-157.
[6].CAÑAVERAS, J.C.; HOYOS, M.;
SÁNCHEZMORAL, S. & SANZ-RUBIO, E.
(1998): The role of microorganisms in underground
karstic environments: examples from Altamira, Tito
Bustillo and Candamo caves (Northern Spain).
[7].CAÑAVERAS, J.C.; GARCÍA DEL CURA,
M.A.; SORIA, J. (EDS.) Sedimentology at the
dawn of the third millenium. 15th International
Sedimentological Congress. Alicante, spain, p. 6-
97.
[8].CAÑAVERAS, J.C.; HOYOS, M.;
SÁNCHEZMORAL, S.; SANZ-RUBIO, E.;
BEDOYA, J.; SOLER, V; LAIZ, L.; GROTH, I.;
SCHUMANN, P.; GONZÁLEZ, I.B.;
SAIZJIMÉNEZ, C. (1999): Microbial communities
associated to hydromagnesite and needlefiber
aragonite deposits in a karstic cave (Altamira,
Spain). Geomicrobiology Journal, nº 16, p. 9-25.
[9].CAÑAVERAS, J.C.; SÁNCHEZ-MORAL, S.;
SOLER, V.; SAIZ-JIMÉNEZ, C. (2001):
Microorganisms and microbially induced fabrics in
cave walls. Geomicrobiology Journal, nº 183,
p.223-240.
[10].CUEZVA, S.; CAÑAVERAS, J.C.;
GONZÁLEZ, R.; LARIO, J.; LUQUE, L.; SÁIZ-
JIMÉNEZ, CUEZVA, S.; CAÑAVERAS, J.C.;
SÁNCHEZ-MORAL, S. (2005):
Biomineralizaciones de huntita en espeleotemas de
la cueva de Altamira (Cantabria). Sem macla, nº 3,
p. 65-66.
[11].CUEZVA, S.; SÁNCHEZ-MORAL, S.,
CAÑAVERAS, J.C.; LARIO, J.; SOLER, V.
(2004): Intercambios de CO2. Suelo/cavidad en un
sistema kárstico somero (cueva de Altamira,
Cantabria). Geotemas, nº 6 (1), p. 327-330.
[12].GENESTE, J.M.; HORDÉ, T. Y TANET, CH
(2003): Lascaux. Une oeuvre de mémoire,
périgueux, p. Fanlac, 142 p.
[13].GENESTE, J.M. (1995): Sous la dir. –
Lascaux, état des lieux, Ministère de la Culture,
Drac Aquitaine, Mai.
[14].GONZALEZ, J.M.; PORTILLO, M.C.;
SAIZJIMENEZ, C. (2006): Metabolically active
crenarchaeota in Altamira cave.
Naturwissenschaften, nº 93, p. 42-45.
[15].GROTH, I., SAIZ-JIMENEZ, C. (1999):
Actinomycetes in hypogean environments.
Geomicrobiology Journal, nº 16, p. 1-8.
[16].Groth, i., vetermann, r., schuetze, b.,
schumann, p., saiz-jimenez, c. (1999):
Actinomycetes in karstic caves of northern Spain
(Altamira and Tito Bustillo). J. Microbiol. Meth., nº
36, p. 115-122.
PROOF
DOI: 10.37394/232020.2022.2.5
Vicente Bayarri-Cayón,
Elena Castillo-López, Jose Antonio Dominguez
E-ISSN: 2732-9941
26
Volume 2, 2022
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the Creative
Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en_US
[17].HERAS MARTÍN, C. ; LASHERAS, J. A.
(2006): L’art paléolithique à Altamira. Revue
monumental, dossier les grottes ornées, semestriel
2, p. 46-49.
[18].LAIZ, L.; GONZALEZ DEL VALLE, M.;
HERMOSIN, B.; ORTIZ MARTINEZ, A.; SAIZ-
JIMENEZ, LAIZ, L.; GROTH, I.; GONZALEZ, I.;
SAIZ-JIMENEZ, C. (1999): Microbiological study
of the dripping waters in Altamira cave (Santillana
del Mar, Spain). J. Microbiol. Meth. nº 36, p. 129-
138
[19].LARIO, J.; SÁNCHEZ-MORAL, S.;
CAÑAVERAS, J.C.; CUEZVA, S.; SOLER, V.
(2005): Radon continuous monitoring in Altamira
Cave (Northern Spain) to assess user’s annual
effective dose. Journal of Environmental
Radioactivity, nº 80, p. 161- 174.
[20].LASHERAS, J.A.; HERAS MARTIN, C.
(2006): Cueva de Altamira and the conservation of
its Palaeolithic art. Coalition: CSIC Thematic
Network on cultural heritage, Electronic
Newsletter, nº 12, p. 7-13. Madrid: CSIC.
Disponible en Web
http:\\www.rtphc.csic.es/boletin
[21].LEROI-GOURHAN, A. (1965): Préhistoire de
l’art occidental, Paris, Citadelles et Mazenod, 485
p. (collection « L'Art et les grandes civilisations ; 1
»)
[22].LEROI-GOURHAN, Ar. y ALLAIN, J.
(1979): Lascaux inconnu, Paris, CNRS Éditions,
381 p. (collection « Supplément à Gallia préhistoire
; 12 »)
[23].MINISTÈRE DE LA CULTURE ET DE LA
COMMUNICATION. France. – État sanitaire de la
cavité. Dossier d'information pour le 32e comité du
patrimoine mondial, Québec, Canada. Paris,
Ministère de la Culture et de la
[24].KHADRA (K.), ANGOT (P.), PAR- NEIX
(S.), CALTAGIRONE (J.-P.). — Fictitious domain
approach for numerical modelling of Navier-Stokes
équations. International journal for numerical
methods in fluids, 34, 2000, p. 651-684.
[25].LACANETTE (D.), MALAURENT (P.),
CALTAGIRONE (J.-P.), VINCENT (S.). — A
model of thermal and aeraulic flows in the cave of
Lascaux. In INTERNATIONAL ASSOCIATION
FOR MATHEMATICAL GEOLOGY —
Quantitative geology from multiple sources: IAMG
06. Congrès international (11; Liège; 2006).
Publication électronique = digital journal.
[26].LACANETTE (D.), MALAURENT (P.),
CALTAGIRONE (J.-P.), BRUNET (J.). — Étude
des transferts de masse et de chaleur dans la grotte
de Lascaux: le suivi climatique et le simulateur.
Karstologia, 50, 2007, p. 19-30.
[27].LACANETTE, D.; VINCENT, S.;
SARTHOU, A.; MALAURENT, P.;
CALTAGIRONE, J.P. An Eulerian/Lagrangian
method for the numerical simulation of
incompressible convection flows interacting with
complex obstacles: Application to the natural
convection in the Lascaux cave. International
Journal of Heat and Mass Transfer vol. 52 issue 11-
12 May, 2009. p. 2528-2542
[28].LACANETTE, D.; LARGE, D.; FERRIER,
C.; AUJOULAT, N.; BASTIAN, F.; DENIS, A.;
JURADO, V.; KERVAZO, BE.; KONIK, S.;
LASTENNET, RO.; MALAURENT, P.; SAIZ-
JIMENEZ, C. A laboratory cave for the study of
wall degradation in rock art caves: an
implementation in the Vézère area. Journal of
Archaeological Science vol. 40 issue 2 February,
2013. p. 894-903
[29].LACANETTE (D.), CALTAGIRONE (J.-P.).
— Le simulateur Lascaux: un outil d'aide à la
décision pour l'avenir de la préhistoire.
Monumental, 2006, 2, p. 94-97.
[30].Larocque, et al. 2010: LAROCQUE, J.;
VINCENT, S.; LACANETTE, D.; LUBIN, P.;
CALTAGIRONE, J.P - Parametric study of LES
subgrid terms in a turbulent phase separation flow..
International Journal of Heat and Fluid Flow vol. 31
issue 4 August, 2010. p. 536-544
[31].MALAURENT (P.), BRUNET (J.),
LACANETTE (D.), CALTAGIRONE (J.- P.). —
Contribution of numerical modelling of
environmental parameters to the conservation of
prehistoric cave paintings: the example of Lascaux
Cave. Conservation and management of
archaeological sites, 8, 2006, p. 1-11.
PROOF
DOI: 10.37394/232020.2022.2.5
Vicente Bayarri-Cayón,
Elena Castillo-López, Jose Antonio Dominguez
E-ISSN: 2732-9941
27
Volume 2, 2022
