See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/321369722

Mapping Chimu's settlements for conservation purposes using UAV and close
range photogrammetry. The virtual reconstruction of Palacio Tschudi, Chan
Chan, Peru

Article  in  Digital Applications in Archaeology and Cultural Heritage · November 2017

DOI: 10.1016/j.daach.2017.11.004

CITATIONS READS

2 172

1 author:

Roberto Pierdicca
Università Politecnica delle Marche
50 PUBLICATIONS   290 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

IEEE Italy Section Consumer Electronics Society Chapter View project

ArcheoFano Project View project

All content following this page was uploaded by Roberto Pierdicca on 29 January 2018.

The user has requested enhancement of the downloaded file.

 Digital Applications in Archaeology and Cultural Heritage xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Digital Applications in Archaeology and Cultural Heritage

journal homepage: www.elsevier.com/locate/daach

Mapping Chimu's settlements for conservation purposes using UAV and close
range photogrammetry. The virtual reconstruction of Palacio Tschudi, Chan
Chan, Peru
Roberto Pierdicca
Università Politecnica delle Marche, Dipartimento di Ingegneria Civile, Edile e Architettura, Italy

A R T I C L E I N F O A B S T R A C T

Keywords: The article deals with the 3D metrical reconstruction of Tschudi Palace, the most famous Palace of Chan Chan
UAV archaeological site, in Peru. The complete mapping of the area was performed by integrating aerial data acquired
Digital photogrammetry with UAV and terrestrial data captured with close range photogrammetry techniques. The input dataset was
Close range acquired during a non-planned survey and with a very cheap equipment. In fact, the acquisitions were made by
Data integration
local and not trained staff. The wind conditions, as well as the presence of visitors made the UAV mission almost
3D modelling
impervious so that the huge part of the work consisted on data filtering and cleaning in order to make them
Virtual reconstruction
Archaeology suitable to be used for photogrammetric purposes. Afterwards, a tidy process of reconstruction was performed,
getting as a final result a thorough ortho-photo of the area. The final check of the model was done with the aid of
sketches made some years before by local experts. The accuracy assessment of the overall product is reported as
well. Once the model was complete, the ground survey of the Plataforma de Entierro, a buried finding conceived
by protective structures, was integrated to the general plan obtained with the first phase. The work demonstrates
how it is possible to achieve good results in terms of accuracy and quality even with coarse data. The present
study would also serve to be an incentive to those researchers who holds priceless datasets of ancient areas and
that should be shared with the research community. As demonstrated, whereas not complete, a 3D survey and
the consequent reconstruction can become paramount not to loose the memory of a place that, for different
reasons, is under continuous hazards.

1. Introduction common condition, even with cheap instruments it is possible to reach

satisfactory results in terms of both accuracy and quality. Moreover, the
The importance of documenting ancient ruins goes beyond the mere growing miniaturization of the components makes UAV more and more
3D representation of an artefact. It means preserving, for the future portable and this is a strong requirement in archaeology, where findings
generations, the memory of a place that, for its fragile nature, is under are difficult to be reached and it is not preferable to have bulky (even
continuous hazards deriving from atmospheric conditions or mankind more accurate) instruments. It is also worth to notice that, nowadays,
negligence. Regardless the cause of these risks, the digitization of pri- aerial vehicles can be equipped with different kinds of sensors, per-
celess architectures becomes essential to secure them and to enable mitting a wide spectrum of applications. And finally, the output pro-
possible reconstructions based on reliable surveys. The techniques ducts that is possible to obtain; Digital Elevation Models (DEM) and
adopted for the documentation in archaeology and cultural heritage ortho-photos (with nadir configuration (Irschara et al., 2010)) and 3D
related projects, should be rapid though accurate, and digital photo- models (using oblique configuration (Jiang et al., 2017)) are clear and
grammetry meets both requirements (Barazzetti et al., 2010; complete, overcoming the limitations of archaeologists, which are still
Remondino et al., 2014). In this context, the advantages of using UAV use to perform their studies with paper drawings. In line with the Se-
(Unmanned Autonomous Vehicles) in archaeology have been proved by ville principles (Bendicho, 2013), 3D virtual models offer the oppor-
several research articles and collected in a recent publication by tunity to enhance the knowledge of cultural goods and should be
Campana (2017). In the article, the author argues about this technology adopted to support the investigations.
from different perspectives. First of all, the costs; though the expenses All the above is supported by recent research trends, which de-
range from few hundreds to hundreds of thousands dollars, for ar- monstrate that the large adoption of aerial platform in archaeology can
chaeological purposes, where the limitation of budgets is a quite serve the following tasks:

E-mail address: r.pierdicca@staff.univpm.it.

https://doi.org/10.1016/j.daach.2017.11.004
Received 1 September 2017; Received in revised form 22 October 2017; Accepted 16 November 2017
2212-0548/ © 2017 Elsevier Ltd. All rights reserved.

Please cite this article as: Pierdicca, R., Digital Applications in Archaeology and Cultural Heritage (2017),
https://doi.org/10.1016/j.daach.2017.11.004
R. Pierdicca Digital Applications in Archaeology and Cultural Heritage xxx (xxxx) xxx–xxx

• Study of ancient civilizations: for the detection of archaeological of the overall product was performed using the statistics of the residuals
features, remote sensing approaches are complementary to the of the photogrammetic block. Secondly, as many buried areas were
ground surveys; in fact, UAVs can act as a virtual eye in the sky conceived by protective structures, a close range photogrammetry
capable to provide substantial information about study areas which survey (even not planned) was integrated to the general plan obtained
otherwise cannot be obtained. By exploiting spectral or thermal with the first phase. The work demonstrates how it is possible to
sensors the study of the evolution of an area can be more detailed achieve good results in terms of accuracy and quality even with coarse
and reliable with respect to a single naked eye approach (Uribe data. UAVs equipped with digital cameras can provide valuable visual
et al., 2015). As well, low weight laser scanners represent an aid to information about the architectures' surface rapidly and at low costs
study the evolutions of the internal organization among different from nearly unstructured data; the research community dealing with 3D
phases of a culture (Sonnemann et al., 2016). And more, by ana- documentation of ancient architectures should be encouraged by this
lysing the morphology of a certain area, taking in consideration the case of study, since there exist a bulk of unstructured data that can be of
slopes and mounds, can help in understanding behavioural patterns use in case of unexpected damages of such structures.
of ancient civilizations and to map them (Asăndulesei, 2017). In the following, after a brief presentation of the case of study of this
• “Emergency” archaeology: it is well known that the destructive work (presented in Section 2), the methodology used to merge UAV and
nature of an excavation can bring to the permanent loss of the site's ground survey is shown in Section 3. The results of the methodology,
original form Malinverni et al. (2016). 3D models and DEMs sig- together with an accuracy evaluation of the method is shown in Section
nificantly contribute to the conservation of archaeological in- 4. Concluding remarks and overall discussion of the work are provided
formation, by documenting the stratigraphy of an excavation. They in Section 5.
allow archaeologists to revisit the site in a virtual space after the
excavation has been concluded (Lonneville et al., 2014). The survey
is realized during the excavations or just at the end of every working 2. Case study
day and drawings have to be produced as soon as possible in order
to allow the comprehension of the work done and to plan the ac- Following previous researches conducted by the author, the work
tivities for the following day. By using this technique, all the mea- described was carried out within the framework of a wider project with
surements, even those not necessary for the day after, have to be the aim of studying Chan Chan site, the capital of the Chimu civilization
acquired in order to avoid a “loss of memory” (Rinaudo et al., 2012). (IX- XV sec.), located 550 km from Lima along the northern coast of
• Documentation: 3D documentation presents an innovative means Peru. The Italian mission of CNR-ITABC in Peru (MIPE) has been
working on the site since 2002 providing the definition of the archae-
of executing and representing measurements taken from archae-
ological sites, objects or contexts. The continuously evolving and ological area and its buffer zone, the archaeological survey of the ter-
improving of sensor technologies, data capture methodologies and ritory, the documentation of the different architectural typologies and
multi-resolution 3D representation can contribute with an important the planning of the archaeological park (Colosi et al., 2013; Colosi and
support to the refinement of information and to the growth of the Orazi, 2015). The territory extends over 14 km2, mainly composed by
archaeological research. It makes possible the acquisition of an ex- great palaces or ciudadelas, elite residences, popular quarters and large
traordinary amount of geo-spatial data, whilst the production of sanctuaries in the shape of stepped pyramids or huacas. The site is
accurate DTM (Haarbrink and Eisenbeiss, 2008) and ortho-mosaic suffering different problems of degradation due to the building mate-
(Themistocleous et al., 2014) permit visual interpretation (thanks to rial, the water rising for capillarity, the centuries-old looting by the
the textures) and hold metrical information. Thus, the reason of the conquistadores and the “huacheros” (grave robbers) and by El Niño.
success of UAV are the very high-resolution of the obtainable pro- Taking into consideration all these degradation problems, the structures
ducts and the availability of easy tools of image processing (Pecci are constantly in hazardous conditions. Becomes then paramount to
and Masini, 2016). digitize with every mean these architectures.
Within this framework, the study case presented is the complete
However, as demonstrated by several examples in the literature mapping of Tschudi Palace, which is also belonging to the Chan Chan
(Pierdicca et al., 2016) for the representation of complex shapes (as site. Built around the 1400 d.C., it is probably the most famous of the
usual in archaeological settings) an “all in one solution” is not existing other ten palaces. It takes his name from the swiss explorer J. Jacobo
so far and the integration of different techniques and sensors proved to Von Tschudi but it is also known as Nik An, since its construction was
be the best solution (Sulaiman et al., 2013). UAV images are often used dedicated to the god Ni, the god of the sea. In this regard, it is richly
in combination with terrestrial surveying in order to close possible 3D embellished with sea and animals decorations (some examples are de-
modelling gaps, mainly due to occlusion issues. Furthermore, there is picted in Fig. 1).
an increasing request and need for digital documentation of archae- In the recent years a great work of restoration was done,1 making
ological sites at different scales and resolutions (Guidi et al., 2008; the Palace the well conserved example of Chimu's architecture existing
Balletti et al., 2015). so far. It is actually the only archaeological site opened to the public in
Given the above, the work described in these pages faces with the Chan Chan, and visible almost for its entirety. The spatial organization
3D reconstruction of a wide archaeological architecture belonging to of the Palace proves a very strict social hierarchy, which can be inferred
the Chan Chan site, in Peru. The problem was tackled from two sides. from a very rigid organization of spaces and borders for the adminis-
First of all, the starting dataset was acquired during a non-planned trative and religious functions. It is composed of three main areas: the
survey and with a very cheap equipment. In fact, the acquisition were north sector which has a “U” shape surrounding a great central square.
made by local and not trained staff. The wind conditions, as well as the The central sector, almost completely occupied by a huge reservoir. The
presence of visitors made the UAV mission almost impervious so that south sector, bordering a second smaller square. The general organi-
the huge part of the work consisted on data filtering and cleaning in zation of the Palace can be seen in Fig. 2.2
order to make them suitable to be used for photogrammetric purposes. The main entrance of the Palace is surrounded by a 10 m height
At the end of this stage, since the flight was not planned, some holes in wall, made of mud brick and adobe; every inner wall have a base of
the final model were present; to overcome this problem, a tidy process around 50 cm (Fig. 1 a). The entrance is simple, situated between the
of reconstruction was done, with the aid of satellite imagery and of
expert archaeologists, getting as a final result a thorough ortho-photo of 1
Restoration works started in 2004 by Instituto Nacional de Cultura La Libertad y la
the area. The final check of the model was done with the aid of sketches Asociación Proyectos para el Desarrollo (PRODE).
2
made some years before by local experts, while the accuracy assessment https://goo.gl/627rAL.

2
R. Pierdicca Digital Applications in Archaeology and Cultural Heritage xxx (xxxx) xxx–xxx

Fig. 1. Pictures of some details of Tschudi

Palace. a) reports a close up view of a typical
corridor interrupted by the walls, b) and c)
depict respectively squirrels and fishes bas
relieves, d) the typical diamond shape with
a clear stylization of a fishing net.

interruption of the walls, which height is accentuated by the thin cor-

ridor the visitor is running. The thickness of the corridors is in contrast
with the dimensions of the Gran Plaza, that is supposed to be com-
pletely painted at the époque of construction. The decorations are im-
pressive, composed of squirrels (Fig. 1 b) to represent the fertility, or in
some cases fishes and diamond pattern which stands to represent
fishing nets (Fig. 1 c and d). In the central sector the visitors' eyes are
captured by the huge reservoir, that was built to overcome the lack of
drinkable water nearby. The south sector is the most important and
holy part of the whole Palace, where was located the grave of the king.
It was located in the center of a large funeral home and surrounded by
44 graves on secondary platforms. Beside his funeral kit were buried
concubines and officiating that would accompany the deceased in life
after death. The burial chamber was built with “T” shape and, at the
Fig. 2. General plan of Tschudi Palace, with highlighted the three sectors. Placed at the time of King's death, the new monarch used to build his own palace
center of Chan Chan, belongs to the ciudadelas Chayhuac y Chol An. The site consisted of over it. The name of the tomb was Plataforma de Entierro and it is visible
large ceremonial and residential spaces, courtrooms, warehouses and the King's tomb. All in Fig. 3. As described in the following, this latter portion of the palace,
in mud and decorated with beautiful relieves. conceived below protective structures, has been georeferenced and
placed in its correct position by integrating the UAV survey and the
ground one.

3. Methodology: integrating aerial and terrestrial data

As stated in the introduction section, the flight campaign was un-

planned and performed with low cost equipment, as well as the ground
survey of the Plataforma de Entierro, which was carried out during a
tourist trip, when local employers were maintaining and protecting the
structures for the upcoming El Niño. This, combined with the vastness
of the Palace (the total coverage of the site is about 444 m × 303 m)
made the 3D reconstruction, starting from such coarse data, a chal-
lenging task. In the following, the steps of the work-flow will be de-
tailed.

Fig. 3. Plataforma de Entierro. 3.1. UAV data acquisition and processing

The pipeline is almost well established and it is composed by the

3
R. Pierdicca Digital Applications in Archaeology and Cultural Heritage xxx (xxxx) xxx–xxx

steps described in Nex and Remondino (2014), summarized as follows:

flight planning and image acquisition, camera calibration and image
orientation, surface reconstruction and orthoimage generation. The
acquisition of stereoscopic images was planned with the aid of Google
satellite images and following the wind conditions; two flights were
planned, one with nadir configuration and one with oblique config-
uration to overcome occlusions. The nadir flight (with an average
height flight of 43 m) was of course made considering the character-
istics of the sensors of the hexa-copter. A DJI S800 drone was used,
installing a Sony Alpha NEX-7 with a resolution of 6000 × 3376 pixel
without the aid of a flight planning software. It has been an important
achievement for the local communities, having a huge echo in the local
press (Industria, 2014).
With an forward overlap of about 60%, was reached the coverage of
the entire area with a total of 1856 pictures. The subject of this case
study is open to the tourists' visit and it was impossible during the flight
to place markers to be used as GCPs, hence the 3D model did not un-
dergone an absolute orientation procedure, exploiting the GNSS or-
ientation of the photogrammetric block; with the aid of direct mea-
sures, a tolerance check of the orthophoto was done; finally, in order to
georeference the projection plane of the orthofoto, visible points from
the satellite image were used. Moreover, the presence of wind gusts
badly affect some pictures, that were automatically analysed by the
software to discard those ones not suitable to be processed. Finally,
1268 images were used. The dataset was redundant but the lack of a
automatic flight control caused the lost of some information. This is
mainly because the images used have no regular overlaps and have
different tilts due to the deleterious influence of the gusts of wind
arising during the flight. After the acquisition phase, the photogram-
metric data followed a traditional procedure of relative orientation with
bundle adjustment using images acquired from the nadir flight. Thanks
to the extreme flexibility of the aerial triangulation algorithm im-
plemented in the photogrammetric software used for the data proces-
sing (Agisoft Photoscan®), it was possible to successfully orient the
images acquired. The flight configuration and the image orientation of
both nadir and oblique acquisitions are depicted in Fig. 4. Given the
huge amount of data it was impossible to compute the triangulation all
at once, hence the photogrammetric model was divided into 15 strips,
computed separately and then joined using visible tie points or, where
not present, using the automatic camera merging.
The alignment of the strips was performed with a tidy process of tie
points recognition, followed by a process of cleaning of the dense point Fig. 4. Flight configuration. a) Depicts the acquisition scheme of the nadir flight, while b)
cloud generated. In fact, since the main objective of the work was to and c) depict the oblique configuration acquisition.
obtain DEM and orthophoto with an average scale of 1:200 and a
Ground Sample Distance (GSD) of 1 cm, all the point cloud has un-
Table 1
dergone a process of filtering and optimization to obtain comparable
Statistics about the flight and about the computation. With an average height flight of
outputs from the different strips. In the following table (Table 1) the 43 m and a 16 mm focal length, a mean GSD of 1 cm was obtained.
average hight flight together with the mean GSD for each chunk are
reported. Strip number N. of images Flight height Focal length GSD (cm/pix)
Even if there were no available GCP to perform an accuracy as-
1 61 40.8 16 0.999
sessment w.r.t the absolute orientation, during the field acquisition, 2 71 40.3 16 1.060
paper drawings and direct measurements were taken, so that it was 3 71 44.3 16 1.085
possible to warrant the suitable accuracy for cartographic purposes and 4 45 42.6 16 1.043
5 71 42 16 1.028
thus preserving the initial GSD. Once recognized the correspondent
5–6 30 39.2 16 0960
points model, the point cloud was optimized according to the real 6 71 40,9 16 1.001
measures (see Fig. 5). 7 72 41.8 16 1.023
The accuracy check values are listed in Table 2, where are reported 7–8 20 40.5 16 0.991
two sample values of the committed error. 8 71 44.7 16 1.094
8–9 30 47.8 16 1.170
It is worth to note that, after the optimization process, the maximum
9 71 42.4 16 1.038
error is comparable with the GSD value, matching with the initial re- 10 71 43.6 16 1.067
quirements of the representation. The final results of this step of the 11 71 43.4 16 1.062
workflow are the DEM and the orthophoto, depicted in Fig. 6. 12 71 43 16 1.053
In the picture is also highlighted with a red circle the position of the 13 71 42.6 16 1.043
14 71 43.1 16 1.055
excavation of the Plataforma the Entierro, that is actually covered by 15 74 42.8 16 1.048
protective structures and that shall be georeferenced in the final model.
In the next subsection, the 3D model of the excavation is reported,

4
R. Pierdicca Digital Applications in Archaeology and Cultural Heritage xxx (xxxx) xxx–xxx

between panoramas was low, impeding a thorough geometrical re-

construction. The procedure was thus switched to a standard Multi
View Stereo processing, using Agisoft Photoscan® to process the single
frames. As in the previous case and for the same reason, GCPs were not
taken during the acquisition; the model was however in an unknown
scale; then, to performed the bundle adjustment, some tie points have
been used taken from the UAV reconstruction. It was in this way pos-
sible to perform the georeferencing of the Plataforma de Entrierro by
replacing it to the protective structures visible from the drone's flight,
keeping advantage from the previous stage. It is worth to notice that,
given the higher GSD of this latter model, it was adapted to the GSD
obtained with the UAV survey. The final 3D model is reported in Fig. 7.

4. Results

In the previous section, the first output of the oriented photo-

grammetric block was showed (being the DEM, afterwards textured
with the relative orthophoto of the Palace). By considering the accuracy
obtained (see Table 1), it is possible to accept a nominal scale ranging
from 1:200 up to 1:100 for the orthophoto produced. Even if the ar-
chaeological site was virtually reproduced, lighting conditions and
some occlusions generated missing information. Holes were manually
closed with the aid of MeshLab,3 choosing the best closing algorithm for
every single hole. This long process was considered for the different
characteristics of the holes related to the dimensions, the position with
respect to the model (flat plane, edge, corner, etc.) and the complexity
of the polygons in the border. An automatic or semi-automatic ap-
proach would have risked neglecting these differences, generating non-
reliable portions in the reality-based model. Fig. 8 is reported with the
purpose of declaring which part of the model arises from the metrical
reconstruction and those one that, due to the lack of information, have
been envisaged to be most close to the reality; the criteria used for this
task was the knowledge of the place and the photographic doc-
umentation available (Fig. 8a). The 3D model undergone a process of
validation by expert archaeologist and, given the assumptions made,
can be considered in compliance with both the London Charter
(Denard, 2012) and the Seville Principles (Bendicho, 2013).
Despite the initial model was composed of more than 100 millions
Fig. 5. Since the relative orientation of the model exploits GNSS coordinates, direct polygons, special care was given to model optimization, in order to
measurement were used to put in scale the final model. a) Input measures within the make the model usable in different platforms and to be managed in 3D
photogrammetry software. b) and c) Sketches of direct measures.
modelling software for the integration with the ground survey (Fig. 8b).
The level of texture resolution was also considered independently by
Table 2 the geometric resolution for maximizing the level of information asso-
Committed error between direct measures and point cloud. The sample measures are ciated with any specific artifact. The final result, obtained with
highlighted in the sketches of Fig. 5.
Blender,4 can be seen in Fig. 9b. It is worthwhile to note that the latest
Measure Distance Error (m) available plan of the Palace dates back to 1974 (Fig. 9a), realized within
the framework of Proyecto Chan Chan Valle de Moche by the University
1 33.40 0.065 of Harvard. As a demonstration of the valuable contribution of the
2 6.02 0.044
computations presented in these pages, in Fig. 9 a comparison among
the old and new general plan of the Palace is depicted.
while in Section 4 the georeferencing process, together with the final
3D modelling and rendering of the entire Palace will be detailed. 5. Discussion and conclusion

The work described in this paper, although exploiting state of art

3.2. Terrestrial reconstruction of the excavation
methodologies, demonstrates how, with a rigorous pipeline, is possible
to achieve an accurate reality-based model (compatible for doc-
3D reconstruction with a textured model of the excavation was
umentation purposes), even starting from unplanned acquisition and
performed using a dataset of images taken from the ground. The da-
coarse datasets. The survey was performed for testing purposes, with
taset, composed of 105 pictures, was captured with a calibrated SONY
low cost equipment but it was able to stem a precious model of a wide
SLT-A77V camera. The camera features are a CMOS Exmor APS-C with
archaeological setting that requires continuous works of restorations.
24,3 MP resolution, coupled with an 24–70 mm objective. The starting
The final products, DEM and orthophotos have been scaled and are
idea was to make the 3D reconstruction with Multi Image Spherical
metrically correct, beside having a sufficient level of details to be used
Photogrammetry ((Fangi, 2011) MISP). For this purpose, several
spherical panoramas were created in order to proceed with the meth-
odology (one of which being visible in Fig. 7). Unfortunately, given the 3
http://www.meshlab.net.
4
complex geometry of the excavation, the number of recognizable points https://www.blender.org.

5
R. Pierdicca Digital Applications in Archaeology and Cultural Heritage xxx (xxxx) xxx–xxx

Fig. 6. Final products of the UAV survey processing. DEM and or-
thophoto are scaled and ready for the following modelling step.

for many purposes. The integration between aerial and ground data to populate an existing web resource of Chan Chan site with the 3D
proved to be a winning solution; in fact, this combination brings the models that our research group is collecting and creating (interested
advantage of georeferencing detailed ground surveys. This is not trivial, readers can find more details in Malinverni et al. (2017a)). The adop-
since in the area the excavations are continuous and the general or- tion of 3D models is becoming also fundamental for in-depth analysis
thophoto paves the way for future findings which can be correctly by the experts. In fact, having at disposal a detailed plan of the area can
placed on the ground, representing a valuable source of information play a pivotal role for a better comprehension of ancient civilizations,
and study for experts archaeologists. as well as for re-designing the visit path of the site. We are already
Regardless the reason of the decay of an archaeological area (hu- working on the integration of the developed model within a dedicated
mans' negligence, fragility of the materials or atmospheric agents), it is Geographical Information System (GIS) (Colosi et al., 2009) that will be
absolutely impossible to predict whether a new damage may occur. The shared with the Peruvian Ministry of Culture; by populating the existing
existence of affordable and reliable pipeline of virtual reconstruction database with further information, local authorities will be enabled for
can represent the turnkey for the safeguard of archaeological goods. By an easier management of the entire Chan Chan archaeological site, fa-
preserving the memory of such places with digital approaches means cilitating also the decision making process. In the future, we also
first of all putting them in safe forever, since owing a faithful 3D model foresee to use this model as a base for implementing change detection
means having the possibility to reconstruct it in the future. Moreover, tasks. In case of future acquisitions in fact, archaeologist can compare
these model can be nowadays easily shared to the mankind thanks to and monitor the changing that may occur at different époques of ac-
the use of different multimedia platforms. In this regards, it is planned quisition (Malinverni et al., 2017b). The present study aims to be an

Fig. 7. 3D model of the Plataforma de

Entrierro. The upper image is one of the
spherical panoramas created, whilst the
lower images depict the 3D reconstruction
from different point of view.

6
R. Pierdicca Digital Applications in Archaeology and Cultural Heritage xxx (xxxx) xxx–xxx

Fig. 8. a) The complete model with highlighted in red the hypothesis

of reconstruction of the missing part. b) Detail of the insertion of the
3D model of the excavation.

Fig. 9. A comparison among the historical

plan available from 1974 by University of
Harvard and the new orthophoto produced
for the presented case of study. A video si-
mulation of the 3D model is available at
https://we.tl/HX0Rq4JSAB.

incentive to those researchers who holds priceless datasets of ancient Appendix A. Supplementary material
areas and that should be shared with the research community. As de-
monstrated, whereas not complete, a 3D survey and the consequent Supplementary data associated with this article can be found in the
reconstruction can become paramount not to loose the memory of a online version at http://dx.doi.org/10.1016/j.daach.2017.11.004.
place.
References

Acknowledgement Asăndulesei, A., 2017. Inside a cucuteni settlement: Remote sensing techniques for doc-
umenting an unexplored eneolithic site from northeastern romania. Remote Sens. 9
(1), 41.
The whole work is carried on with the contribution of the Italian Balletti, C., Guerra, F., Scocca, V., Gottardi, C., 2015. 3d integrated methodologies for the
Ministry of Foreign Affairs and in a strict collaboration with the documentation and the virtual reconstruction of an archaeological site. Int. Arch.
Ministerio de Cultura and the Direccin Desconcentrada de Cultura de La Photogramm. Remote Sens. Spat. Inf. Sci. 40 (5), 215.
Barazzetti, L., Remondino, F., Scaioni, M., 2010. Automation in 3d reconstruction: results
Libertad. I would like to thank arch. Roberto Orazi (scientific director of on different kinds of close-range blocks. In: Proceedings of the ISPRS Commission V
the MIPE project) e Francesca Colosi from ITABC CNR in Rome, Italy, Symposium International Archives of Photogrammetry, Remote Sensing and Spatial
that for years are lavishing their efforts to promote the knowledge of Information Sciences. Newcastle upon Tyne, UK, vol. 38.
Bendicho, V.M.L.-M., 2013. International guidelines for virtual archaeology: the seville
Chan Chan site. I'm also glad to the local staff for the support and for the principles. In: Good Practice in Archaeological Diagnostics. Springer, pp. 269–283.
dataset released to perform the 3D reconstruction. Is my wish to thank Campana, S., 2017. Drones in archaeology. state-of-the-art and future perspectives.
Giacomo Gattoni who was in charge of the data processing, our stu- Archaeol. Prospect (n/a-n/aARP-16-0037.R1). http://dx.doi.org/10.1002/arp.1569.
Colosi, F., Fangi, G., Gabrielli, R., Orazi, R., Angelini, A., Bozzi, C.A., 2009. Planning the
dents, our greatest resource. And finally, I thank professor Eva Savina
archaeological park of chan chan (peru) by means of satellite images, gis and pho-
Malinverni for improving the article's quality with her suggestions but, togrammetry. J. Cult. Heritage 10, e27–e34.
foremost, for trusting me. Colosi, F., Gabrielli, R., Orazi, R., Malinverni, E.S., 2013. Discovering chan chan: modern
technologies for urban and architectural analysis. Archeologia e Calcolatori 24,
187–207.

7
R. Pierdicca Digital Applications in Archaeology and Cultural Heritage xxx (xxxx) xxx–xxx

Colosi, F., Orazi, R., 2015. Project for the archaeological park of chan chan, Peru. In: Malinverni, E.S., Pierdicca, R., Colosi, F., Orazi, R., 2017a. Web tool as a virtual museum
Heritage and Technology. Mind, knowledge, Experience. Le Vie dei Mercanti, XIII of ancient archaeological ruins in peru. In: International Conference on Augmented
Forum Internazionale di Studi. Aversa, pp. 2031–2041. Reality, Virtual Reality and Computer Graphics. Springer, pp. 282–296.
Denard, H., 2012. A new introduction to the london charter. Paradata Transpar. Virtual Nex, F., Remondino, F., 2014. Uav for 3d mapping applications: a review. Appl. Geomat.
Heritage 57–72. 6 (1), 1–15.
Fangi, G., 2011. The Multi-image Spherical Panoramas as a Tool for Architectural Survey. Pecci, A., Masini, N., 2016. Archaeology, historical site risk assessment and monitoring by
Cipa Heritage Documentation, Best Practices and Applications XXXVIII-5/C19, 21. uav: approaches and case studies. In: EGU General Assembly Conference Abstracts,
Guidi, G., Remondino, F., Russo, M., Menna, F., Rizzi, A., 2008. 3d modeling of large and vol. 18, p. 17424.
complex site using multi-sensor integration and multi-resolution data. In: VAST, pp. Pierdicca, R., Frontoni, E., Malinverni, E.S., Colosi, F., Orazi, R., 2016. Virtual re-
85–92. construction of archaeological heritage using a combination of photogrammetric
Haarbrink, R., Eisenbeiss, H., 2008. Accurate dsm production from unmanned helicopter techniques: Huaca arco iris, chan chan, peru. Digit. Appl. Archaeol. Cult. Heritage 3
systems. Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. 37, 1259–1264. (3), 80–90.
Industria, L., Thuersday, 7th August 2014 Trujillo, La Libertad. La grandeza de chan chan Remondino, F., Spera, M.G., Nocerino, E., Menna, F., Nex, F., 2014. State of the art in
vista desde las alturas. In: Un dron registra ineditas tomas del complejo arqueologico high density image matching. Photogramm. Rec. 29 (146), 144–166.
Chimu. Director de Proyecto Especial Henry Gayoso Pareded, destaca aplication de Rinaudo, F., Chiabrando, F., Lingua, A.M., Spanò, A.T., 2012. Archaeological site mon-
dispositivo. Redaccion La Industria, p. A.12. itoring: Uav photogrammetry can be an answer. Int. Arch. Photogramm. Remote
Irschara, A., Kaufmann, V., Klopschitz, M., Bischof, H., Leberl, F., 2010. Towards fully Sens. Spat. Inf. Sci. 39 (B5), 583–588.
automatic photogrammetric reconstruction using digital images taken from uavs. Sonnemann, T.F., Ulloa Hung, J., Hofman, C.L., 2016. Mapping indigenous settlement
Jiang, S., Jiang, W., Huang, W., Yang, L., 2017. Uav-based oblique photogrammetry for topography in the caribbean using drones. Remote Sens. 8 (10), 791.
outdoor data acquisition and offsite visual inspection of transmission line. Remote Sulaiman, N.S., Bachad, E.W., Chong, A.K., Majid, Z., Setan, H., 2013. Generating three-
Sens. 9 (3), 278. dimensional photo-realistic model of archaeological monument via multi-sensorsdata
Lonneville, B., De Roo, B., Stal, C., De Wit, B., De Wulf, A., De Maeyer, P., 2014. Accurate integration. Int. J. Res. Eng. Technol. 2 (10), 362–373.
and cost-efficient 3d modelling using motorized hexacopter, helium balloons and Themistocleous, K., Agapiou, A., King, H.M., King, N., Hadjimitsis, D. G., 2014. More than
photo modelling: A case study. In: Euro-Mediterranean Conference. Springer, pp. a flight: the extensive contributions of uav flights to archaeological research-the case
410–417. study of curium site in cyprus. In: Euro-Mediterranean Conference. Springer, pp.
Malinverni, E., Barbaro, C.C., Pierdicca, R., Bozzi, C., Tassetti, A., 2016. Uav surveying for 396–409.
a complete mapping and documentation of archaeological findings. the early neo- Uribe, P., Angás, J., Pérez-Cabello, F., de la Riva, J., Bea, M., Serreta, A., Magallón, M.A.,
lithic site of portonovo. Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. 41. Sáenz, C., Martín-Bueno, M., 2015. Aerial mapping and multi-sensors approaches
Malinverni, E.S., Pierdicca, R., Sturari, M., Colosi, F., Orazi, R., 2017b. Documentation from remote sensing applied to the roman archaeological heritage. Int. Arch.
and detection of colour changes of bas relieves using close range photogrammetry. Photogramm. Remote Sens. Spat. Inf. Sci. 40 (5), 461.
Int. Arch. Photogramm., Remote Sens. Spat. Inf. Sci. 42.

View publication stats