buildings
Article
Multi-Analytical Investigations of the Medieval Turkish Bath
from Goles, ti Open Air Museum
Monica Dinu 1, * , Lucian Cristian Ratoiu 1 , Camelia Călin 2 and Gerard Călin 2
Citation: Dinu, M.; Ratoiu, L.C.;
Călin, C.; Călin, G. Multi-Analytical
Investigations of the Medieval
Turkish Bath from Goles, ti Open Air
Museum. Buildings 2023, 13, 321.
https://doi.org/10.3390/
buildings13020321
Academic Editor: Eva Barreira
Received: 2 December 2022
Revised: 17 January 2023
Accepted: 18 January 2023
Published: 21 January 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Buildings 2023, 13, 321. https://doi.org/10.3390/buildings13020321
1 National Institute of Research and Development for Optoelectronics INOE 2000, 077125 Măgurele, Romania
2 Goleşti Museum, 117717 Golesti, Romania
* Correspondence: monica.dinu@inoe.ro
Abstract: The Goles, ti Manor ensemble consists of architectural monuments dating back to the 17th,
18th, and 19th centuries, with ephemeral architecture roots from Phanariot times but also some pre-
Brancovan influences. The Turkish steam bath is placed in the northeast of the enclosure, and it is
thought to have been built by Stroe Leurdeanu. Although there are still original fragments of the old
plaster that have been preserved, the Turkish bath has undergone several restorations through time,
some of them poorly documented. Hyperspectral imaging (HSI) combined with LIBS were used
in order to analyze the bricks and the mortars, evaluate their preservation state, map the original
elements, and document and classify the interventions previously made. Based on the supervised
classifications, several areas of interest were selected for LIBS elemental analysis, and multivariate
data analysis was conducted in order to discriminate similar composition materials and to map
the interventions and the original. By analyzing the variability of the spectral profiles, different
algorithms were applied to the SWIR hyperspectral images in order to classify the main elements of
the brickwork.
Keywords: medieval bricks; Turkish bath; LIBS; hyperspectral imaging
1. Introduction
The subject of the current study is a building made from clay bricks and mortars: the
Turkish bath from the Goles, ti Manor estate. Clay brick structures, with or without mortars,
represent one of the oldest and sturdy techniques used since the Mesopotamian, Egyptian,
and Roman times until medieval and modern times. Although technology advanced
through the ages, bricks continued to have a simple manufacturing process based on clay,
a raw material available in large quantities all over the Earth. There are a lot of brick
masonries that lasted through centuries of thaw–freezing cycles, high temperatures, and
anthropogenic factors that prove that it is an effective building material that provides both
resistance to prevalent climatic conditions and insulation from cold and heat [1].
In the 17th century, the feudal ensemble of Goles, ti (Arges, County) was built around
two major constructions: the manor built in 1640 and the church erected six years later. For
200 years, the main buildings and annexes underwent major changes, but only a small part
of them can be documented with the existing information [2].
In the northwest corner of the enclosure is located the only part of the building that
was once the Turkish bath. The only documentary mention of the bathing custom at the
Goles, ti mansion was written by Paul of Aleppo and is dated 12 January 1657. Speaking
about the house built by the great boyar Stroe Leurdeanu in 1640 on his rural estate, the
Syrian deacon says: “His palace is big, with many baths” [3]. Built according to the Ottoman
model, which in turn inherited the water culture from the Roman–Byzantine area, the hot
baths (hammams) have been present in Romanian cultural spaces since the middle of the
16th century [4]. Constantin Cantacuzino, Stroe Leurdeanu’s political rival, and also a great
and powerful boyar, had baths both in his house in Bucharest and in his aristocratic estate
https://www.mdpi.com/journal/buildings
Buildings 2023, 13, x FOR PEER REVIEW 2 of 12

Buildings 2023, 13, 321 2 of 11

of the 16th century [4]. Constantin Cantacuzino, Stroe Leurdeanu’s political rival, and also
a great and powerful boyar, had baths both in his house in Bucharest and in his aristocratic
estate from Filipești (Prahova County), as Paul of Alep himself testifies: ‘an elegant bath
from Filipes
whose , ti (Prahova
marble County),
is marvelous; as water
the Paul ofthat
Alepfeeds
himself testifies:
it is brought ‘an by
elegant bathwith
wheels whosebuckets
marble is marvelous; the water that feeds it is brought by wheels with buckets placed on
placed on the river; these wheels also sprinkle the vegetable gardens and the beautiful
the river; these wheels also sprinkle the vegetable gardens and the beautiful orchards; they
orchards; they set many mills in motion’ [3]. The bath was not only hygienic, but a place
set many mills in motion’ [3]. The bath was not only hygienic, but a place of socializing
of socializing
and relaxation,and relaxation,
where men andwhere
womenmenspentand women
hours spent hours of delight.
of delight.
In
In the interwarperiod,
the interwar period,the
theGolescu
Golescu manor
manor waswas almost
almost in ruins;
in ruins; the outbuildings
the outbuildings had had
collapsed, and the vaults of the Turkish bath were about to fall down, as
collapsed, and the vaults of the Turkish bath were about to fall down, as can be seen incan be seen in
Figure
Figure 1a. During the 1942–1943 restoration, only one room was preserved from the Turkish Turk-
1a. During the 1942–1943 restoration, only one room was preserved from the
ish bath:
bath: the the
warm warm space
space with with a single
a single dome dome
and the and the adjacent
adjacent furnace.furnace.

(a) (b)
Figure
Figure 1. The outbuildings
outbuildingsfeaturing
featuring
thethe Turkish
Turkish bathbath in 1942
in 1942 (a) at
(a) and and at present
present time (b).
time (b).

The casuistic
The casuistic under
underdiscussion
discussion is is
focused
focusedon on
thethe
discrimination of the
discrimination oforiginal mate-mate-
the original
rials, bricks, and mortars, considering that, during the restoration, new bricks were also
rials, bricks, and mortars, considering that, during the restoration, new bricks were also
incorporated in addition to the original remains. Another question was if the mortars
incorporated in addition to the original remains. Another question was if the mortars used
used in the restoration also contain shards from the original bricks. In such a compli-
in the restoration
cated also we
situation where contain
need shards from thebetween
to discriminate originalvery
bricks. In such
similar a complicated
objects, statistical situ-
ation where we need
and classification to discriminate
methods are the only between very Thus,
solution [5,6]. similar objects, statistical
hyperspectral analysisand classifi-
(HSI)
cation
and laser-induced breakdown spectroscopy (LIBS) were selected for their great classifica- laser-
methods are the only solution [5,6]. Thus, hyperspectral analysis (HSI) and
induced
tion and breakdown
chemometricspectroscopy (LIBS)
capabilities [7], and awere selected
charting of thefor their bricks
original great and
classification
mortars and
chemometric capabilities [7], and a charting of the original bricks and mortars was ob-
was obtained [8].
tained [8].
2. Materials and Methods
In the form
2. Materials andin which it was restored, as can be seen in Figure 1b, the Turkish bath
Methods
has the following characteristics: a rectangular shape with sides of 8 m/6.3 m and a 6.4 m
In the form in which it was restored, as can be seen in Figure 1b, the Turkish bath has
height, thick walls of about 0.8 m built of bricks bonded with thick layers of mortar. The
the following
bath room itself characteristics:
is square in shapea rectangular
with a sideshape with
of about 5.5sides
m. Theof 8ceiling,
m/6.3 also
m andmade a 6.4 m
height,
of bricks,thick wallsinof
is built about 0.8 m built
a semicylindrical of bricks
vault piercedbonded with thick
by openings layers
for excess of mortar.
steam. The The
bath
lime room
and sand itself is square
plaster were in shape
mixed withwith a side
crushed of about
brick, giving5.5 them. The ceiling,
interior a reddishalso made of
color.
bricks,
On the is built init ahas
outside, semicylindrical
a profiled cornice vault pierced
made by openings
of several successive forlayers
excessof steam.
brick. TheThe lime
ornamentation
and sand plaster features
wereamixed
saw tooth
withor crushed
wolf’s tooth motif
brick, [9]. Inthe
giving thisinterior
paper, discussions will On
a reddish color.
envisage
the outside,an area from
it has the mostcornice
a profiled representative
made ofpart of the historical
several successive building
layersinofterms
brick.ofThe
age orna-
value that is alleged to contain original bricks and mortars, as well as materials
mentation features a saw tooth or wolf’s tooth motif [9]. In this paper, discussions will from later
restoration interventions. In the middle of the investigated area, there is a brick that stands
envisage an area from the most representative part of the historical building in terms of
out due to its aspect and its length, which is known to be part of a restoration.
age value that is alleged to contain original bricks and mortars, as well as materials from
As the properties of medieval and modern bricks vary in terms of raw materials,
later restoration
production interventions.
methodology, as wellInas
the middleused,
mortars of the
hisinvestigated
combined with area,LIBS
there is aused
were brick that
stands
in order outtodue to itsthem,
analyze aspect and itstheir
evaluate length, which is known
preservation state, map to be part
the of a restoration.
original elements,
As the properties of medieval and modern bricks vary
and document and classify the interventions previously made [10,11]. The main in terms of raw materials,
chemical pro-
duction methodology,
constituents of clay are as well (Si)
silicon as mortars used, his
and aluminum combined
(Al), but alsowith LIBS were
a number used
of minor in order
and
to analyze
trace elementsthem,thatevaluate theirinpreservation
will be used state,
multivariate data map the
analysis original
in order elements, and
to discriminate and docu-
map the
ment andoriginal
classifymaterials.
the interventions previously made [10,11]. The main chemical constitu-
ents of clay are silicon (Si) and aluminum (Al), but also a number of minor and trace
 wavelength λ and spatial (x, y) dimensions. The results of the hyperspectral analysis can
be used either as spectral data I(λ) for each individual pixel (x, y) or as an image I(x, y) for
each wavelength λ [13]. Each pixel is attributed a unique spectral fingerprint, which can
be used to characterize its chemical composition. As such, hyperspectral imaging can be
Buildings 2023, 13, 321
used for the identification and quantification of chemical compositions and, at the same
3 of 11
time, for mapping their distribution [14].
For each pixel in an image, a hyperspectral camera acquires the light intensity (radi-
ance) for a large number of contiguous spectral bands; every pixel in the image thus in-
Hyperspectral imaging or imaging spectroscopy is a noninvasive, noncontact tech-
cludes a continuous spectrum (in radiance) and can be used in order to characterize the
nique which combines digital imaging and spectroscopy [12]. HSI offers both imaging and
objects in the scene with great precision and detail. For the HSI analyses, a HySpex SWIR-
spectral data, measuring the spatial information and the spectral parameters for each pixel
384
simultaneously. Thecamera
hyperspectral from NEO
hyperspectral imageequipped
is expressedwith
as aahypercube
state-of-the-art Mercury–Cad-
I(x, y, λ), containing
mium–Telluride (MCT) detector was used, calibrated for the following
wavelength λ and spatial (x, y) dimensions. The results of the hyperspectral analysis lenses: 0.3can
m,be 3 m,
9used
m, and foras
either a field of view
spectral data (FOV)
I(λ) forof 16 degrees.
each individualThe system
pixel (x, y)covers
or as anparts of the
image I(x,NIR
y) forand
SWIR spectra from 954 nm to 2550 nm and has 288 spectral channels. Taking
each wavelength λ [13]. Each pixel is attributed a unique spectral fingerprint, which can be into consid-
eration
used to the availableitsworking
characterize chemicalconditions
composition. on-site (very
As such, narrow access,
hyperspectral imagingsee can
Figure 2a but
be used
also theidentification
for the need for theand bestquantification
available resolution,
of chemicala 30 cm distance
compositions close-range
and, at the samelenstime,was
forin-
stalled.
mapping It was
theirhighly important
distribution [14]. to record from close range in order to obtain greater details
Fortoeach
but also avoidpixel in anmixing
spectral image, of a hyperspectral
the pixel signalscamera
(whenacquires the light
pixels become intensity
larger in size(ra-
with
diance) for a large number of contiguous spectral bands; every pixel in
the increase in distance, the spectral profile can combine the specific signals of different the image thus
includes
types a continuous spectrum (in radiance) and can be used in order to characterize
of materials).
the objects
Data was in the scene with
recorded using great
the precision
push-broom andtechnique
detail. For(athe HSI analyses,
long-track a HySpex
scanner meaning
SWIR-384 hyperspectral camera from NEO equipped with a state-of-the-art
successive lines of 384 pixels, which represent the width of a recording) by scanning the Mercury–
Cadmium–Telluride (MCT) detector was used, calibrated for the following lenses: 0.3 m,
vertical surface using a motorized translation stage.
3 m, 9 m, and for a field of view (FOV) of 16 degrees. The system covers parts of the NIR
During the investigation, diffuse illumination was provided by two custom-made
and SWIR spectra from 954 nm to 2550 nm and has 288 spectral channels. Taking into
lamps, which focused the illumination to a line overlapping with the camera FOV. The
consideration the available working conditions on-site (very narrow access, see Figure 2a
lens ensured a FOV (field of view) of 16 degrees, meaning an 84 mm width for each ac-
but also the need for the best available resolution, a 30 cm distance close-range lens was
quisition.
installed. For the highly
It was recording of the interest
important to recordarea,
from7 acquisitions
close range in were
ordernecessary.
to obtainProcessing
greater
was conducted using ENVI ® L3Harris Geospatial. All data were converted
details but also to avoid spectral mixing of the pixel signals (when pixels become larger from DN (dig-
in
ital number) to radiance units and were later radiometrically calibrated
size with the increase in distance, the spectral profile can combine the specific signals using QUAC
of
(quick atmospheric
different correction).
types of materials).

Camera

TS
controller

(a) (b)
Figure
Figure 2.
2. HSI
HSI system
system (a) and
and LIBS
LIBS(b)
(b)data
dataacquisition
acquisitionononthe
the
NENE side
side of of
thethe Turkish
Turkish bath.
bath.

Data is
LIBS was recorded usingtechnique,
a microinvasive the push-broom technique
and each (a long-track
triggered scannerameaning
pulse generates spectrum,
successive lines of 384 pixels, which represent the width of a recording) by scanning the
advancing deeply in the layers, pulse by pulse, until it reaches the original material
vertical surface using a motorized translation stage.
During the investigation, diffuse illumination was provided by two custom-made
lamps, which focused the illumination to a line overlapping with the camera FOV. The
lens ensured a FOV (field of view) of 16 degrees, meaning an 84 mm width for each
acquisition. For the recording of the interest area, 7 acquisitions were necessary. Processing
was conducted using ENVI® L3Harris Geospatial. All data were converted from DN
(digital number) to radiance units and were later radiometrically calibrated using QUAC
(quick atmospheric correction).
LIBS is a microinvasive technique, and each triggered pulse generates a spectrum,
advancing deeply in the layers, pulse by pulse, until it reaches the original material unaf-
fected by adherent deposits, corrosion, chemical changes, etc. [15]. The main feature of the
technique is the fact that each laser pulse delivers information about the chemical elements
present in the ablated material, providing a sort of cross-sectional analysis until it reaches
Buildings 2023, 13, 321 4 of 11

the core of the material [16,17]. Thus, the major advantage that the LIBS technique has
over other methods of analysis currently used in the field of investigations (e.g., XRF, FTIR,
XRD) is determined by the possibility to perform in situ stratigraphic analyses without
sampling the object (without processing or even cleaning of surfaces) [18–21]. LIBS analy-
ses were made using a handheld spectrometer from SciAps that works in an argon purge
environment, see Figure 2b. The equipment has a Q-switched Nd:YAG emitting at 1064 nm
with 5 mJ of energy and a 50 µm laser spot. Its spectrometers cover a spectral range from
190 nm to 950 nm. Due to the fact that the bricks are a heterogenous materials, the LIBS
acquisitions were made using matrix features of the equipment for 3 × 3 to 6 × 6 spots and
20 to 100 series, depending on the area investigated (mortar, brick, adherent deposits).
In order to discriminate the original materials as well as historical and current inter-
ventions in the brickwork, the LIBS data were subjected to principal component analysis
(PCA) conducted using Origin Lab. Afterwards, the hyperspectral and LIBS PCA results
were correlated and, further on, processed in ENVI using spectral angle mapper (SAM)
and linear spectral unmixing (LSU) algorithms in order to classify and trace the different
areas of interest [22]. Both algorithms were applied complementary in order to validate
the accuracy of the classifications and mapping [23,24]. SAM is a supervised spectral
classification algorithm that uses the difference between angles in an n-dimensional space
to classify pixels in an image [25]. It determines the similarity degree from spectral pro-
files by calculating the angle between the spectra while interpreting them as vectors in
an n-dimensional space, where n equals the number of bands. Smaller angles indicate a
higher degree of similarity to the reference spectrum, while larger angles designate a lower
degree of similarity. Pixels with the smallest angle values are associated (grouped) to the
same class of materials. Due to the fact that the SAM algorithm is part of the supervised
classification, the manual selection of the endmembers represents a key feature for class
assignment [26]. ENVI enables two types of threshold value selections: (1) a single value
for all classes and (2) a customized value for every type of class. The value of 0.100 radians
is the default value of the menu, and it means that a group of pixels is assigned to a class if
the difference between them is not greater than 0.100 radians (in SAM, pixels are considered
as vectors in an n-dimensional space). The spectral profile selected for each class is different
enough in order to enable the classification (spectral variability) since several values were
used, and no major variations in the results were observed.
LSU is a method able to generate abundance maps of pixels with specific spectral
profiles based on the pixel purity characteristics [27]. This is made by identifying the
endmembers in the scene and the fractional abundances that correspond to each pixel
category. Moreover, LSU calculations allow for the classification of areas with mixed pixels
based on their statistical values.

3. Results and Discussions

3.1. HSI Analysis
The different behavior of materials at various wavelengths can be observed as sig-
nificant changes in the absorbance and reflectance of the NIR and SWIR radiation on the
surface through the 954–2550 nm spectral domain [28].
Raw hyperspectral data were expressed in digital numbers (DN) and were made avail-
able via HySpex GROUND. Using HySpex RAD, the raw data were converted into radiance
units. HySpex RAD is a standalone postprocessing software facilitating radiometric cali-
bration of HySpex images. HySpex RAD enables both raw (_raw) and real-time-calibrated
(_corr) images to be converted to at-sensor absolute radiance values (W/sr·nm·m2 ). The
output files were stored in BSQ format and were converted to 32 bits, which do not need a
scaling factor to convert the images to radiance. A second radiometric calibration of the
radiance data was made using the QUACK module (quick atmospheric correction) of ENVI
in order to minimize the influence of the air and light [29]. This atmospheric correction
is based on the empirical finding that the average reflectance of diverse material spectra
(excluding highly structured materials, such as vegetation, shallow water, and mud) is not
 which
which dodonot
notneed
needa ascaling
scalingfactor
factortotoconvert
convertthe theimages
imagestotoradiance.
radiance.AAsecond
secondradiometric
radiometric
calibration
calibrationofofthetheradiance
radiancedatadatawas wasmademadeusing
usingthe theQUACK
QUACKmodule module(quick
(quickatmospheric
atmospheric
correction)
correction)ofofENVIENVIininorder
ordertotominimize
minimizethe theinfluence
influenceofofthe theair
airand
andlight
light[29].
[29].This
Thisat-at-
Buildings 2023, 13, 321
mospheric
mosphericcorrection
correctionisisbased
basedon onthe
theempirical
empiricalfinding
findingthatthatthe
theaverage
averagereflectance
reflectanceof of di-
di-
5 of 11
verse
versematerial
materialspectra
spectra(excluding
(excludinghighlyhighlystructured
structuredmaterials,
materials,such
suchasasvegetation,
vegetation,shallow
shallow
water,
water,and andmud)
mud)isisnotnotdependent
dependenton oneach
eachscene.
scene.TheTheoutput
outputafter
aftercorrection
correctionisisexpressed
expressed
asasapparent
apparentreflectance
reflectanceinteger
integerdata,
data,where
wherepixel
pixelvalues
valuesrange
rangefrom
from00toto10,000
10,000(represent-
(represent-
dependent on each scene. The output after correction is expressed as apparent reflectance
inging0 0toto100%
100%reflectance).
reflectance).
integer data, where pixel values range from 0 to 10,000 (representing 0 to 100% reflectance).
Figure
Figure 3a,bpresents
3a,b presentsrelevant
relevantimages
imagesacquired
acquiredfor for954
954andand2200
2200nmnmwavelengths,
wavelengths,and and
Figure 3a,b presents relevant images acquired for 954 and 2200 nm wavelengths, and
absorption,
absorption,reflectance,
reflectance,andandtransmittance
transmittancedifferences
differencescan canbe beobserved.
observed.As Asthe
thewavelength
wavelength
absorption, reflectance, and transmittance differences can be observed. As the wavelength
increases,
increases,somesomeinsight
insighton onthe
thebricks
bricksprevails.
prevails.Using
Usingthe thefalse
falsecolour
colourinfrared
infrared(FCIR)
(FCIR)func-
func-
increases, some insight on the bricks prevails. Using the false colour infrared (FCIR)
tion,
tion,RGBRGBchannels
channelswerewereassigned
assignedtotodifferent
differentcombinations
combinationsofofwavelengths
wavelengthsranging
rangingfromfrom
function, RGB channels were assigned to different combinations of wavelengths ranging
954954to 2250
to nm
2250 nmininorder
order to track
to track the
the similarities
similarities and
and differences
differences in
from 954 to 2250 nm in order to track the similarities and differences in the material inthe
the material
material behaviour.
behaviour.
Figure3c3cpresents
Figure
behaviour. presents
Figurethe the
3c FCIRfor
FCIR
presents for
thethe
the RGBfor
RGB
FCIR channels
channels
the RGB attributed
channelstoto
attributed 954,1200,
954, 1200,
attributed toand
and954, 2000
2000 nm
nm
1200, andin
in
which
which one
one can
can observe
observe three
three colours
colours for
for bricks
bricks and
and mortars
mortars that
that are
are
2000 nm in which one can observe three colours for bricks and mortars that are used further used
used further
further on
on for
for
endmember
endmember
on selection.
selection.
for endmember selection.

(a)
(a) (b)
(b)

(c)
(c)
Figure
Figure
Figure 3.
3.3. HSI
HSI imagesfor
images
HSI images for(a)
for (a)954
(a) 954nm,
954 nm,(b)
nm, (b)2200
(b) 2200nm,
2200 nm,and
nm, and(c)
and (c)FCIR.
(c) FCIR.
FCIR.

3.2. LIBS
3.2.LIBS
3.2. LIBS
The areas
Theareas
The for
areasfor LIBS
forLIBS stratigraphy
LIBSstratigraphy were
stratigraphywere selected
wereselected based
selectedbased on
basedon the
onthe hyperspectral
thehyperspectral imaging
hyperspectralimaging
imaging
results, and
results,and
results, they
andthey are
theyare presented
arepresented in Figure
presentedininFigure 4.
Figure4.4.

Figure4.
Figure 4. LIBS
LIBS analysis
analysis spots.
spots.
Figure 4. LIBS analysis spots.

The
Themain
main chemical
chemical elements
elementsidentified
identifiedusing
usingLIBS,
LIBS,presented
presentedin inTable
Table1,1,can
canbe betraced
traced
The main chemical elements identified using LIBS, presented in Table 1, can be traced
to
to oxides
oxides commonly
commonly found
foundin inclay
claybricks
bricksasasfollows:
follows:silica
silica(SiO
(SiO2 ),
2),alumina
alumina(Al (Al2 2O
O33),), iron
iron
to oxides commonly found in clay bricks as follows: silica (SiO2), alumina (Al2O3), iron
(Fe2 O3 ) or ferrous oxide (Fe3 O4 ), potassium oxide (K2 O), titanium dioxide (TiO2 ), sodium
oxide (Na2 O), calcium oxide (CaO), and magnesium oxide (MgO), as well as other elements
such as barium (Ba), strontium (Sr), rubidium (Rb), and manganese (Mn) [1].
 ments such as barium (Ba), strontium (Sr), rubidium (Rb), and manganese (Mn) [1].
Buildings 2023, 13, x FOR PEER REVIEW 6 of 12
Table 1. Chemical elements identified using LIBS.

Type
Buildings 2023, of
13, 321 Material Main Chemical Elements Identified by LIBS Averaged Data 6 of 11
(Fe2O3) or ferrous oxide (Fe3O4), potassium oxide (K2O), titanium dioxide (TiO2), sodium
Bricks oxide (Na2O),Si,calcium
Al, Mg,oxide
Fe, K, Na, and
(CaO), Ca,magnesium
O, C, Ti oxide (MgO), as well as other ele-
Reddish mortarments
areassuch asCa, Si, Sr,
barium Al,strontium
(Ba), Na, Mg,(Sr),
K,rubidium
C, Fe, O,(Rb),
Ti and manganese (Mn) [1].
Table 1. Chemical elements identified using LIBS.
Grey mortar areas Ca, Si, Na, Al, K, C, Sr, O
Table 1. Chemical
Type elements identifiedMain
of Material usingChemical
LIBS. Elements Identified by LIBS Averaged Data
Type of Material
Bricks Main Chemical Elements Identified by LIBS Averaged Data
Si, Al, Mg, Fe, K, Na, Ca, O, C, Ti
The spots where LIBS spectra were acquired were carefully selected using the mag-
Bricks
Reddish mortar areas Si, Al, Mg,
Ca,Fe,Si, K,
Sr, Na, Ca,Mg,
Al, Na, O, C, Ti Fe, O, Ti
K, C,
nifying camera of the system to focus on similar structure areas for the bricks and the
Reddish mortarareas
Grey mortar areas Ca, Si, Sr,Ca,
Al,Si,Na,
Na,Mg, K,C,C,Sr,Fe,
Al, K, O O, Ti
mortars, respectively. Figure 5 presents
Grey mortar areas
the averaged signals obtained for the Si, Al, and
Ca, Si, Na, Al, K, C, Sr, O
Ca lines for the whole batch.
The spots where LIBS spectra were acquired were carefully selected using the magni-
The spots where LIBS spectra were acquired were carefully selected using the mag-
fying camera of the system to focus on similar structure areas for the bricks and the mortars,
nifying camera of the system to focus on similar structure areas for the bricks and the
respectively. Figure 5 presents the averaged signals obtained for the Si, Al, and Ca lines for
mortars, respectively. Figure 5 presents the averaged signals obtained for the Si, Al, and
the whole batch.
Ca lines for the whole batch.

(a) (b)
Figure 5. LIBS spectra for (a) Si line (a)
and (b) Ca and Al lines. (b)
Figure 5. LIBS
Figure spectra
5. LIBS forfor
spectra (a)(a)
Si line and
Si line (b)(b)
and CaCa
and Al Al
and lines.
lines.
PCA was applied to the LIBS data to help classify and discriminate them. Taking into
consideration previous
PCAPCA
waswas
studies applied
related
applied to to
to the the
LIBS LIBS
the data to help
statistical
data to help classify
analysis
classify and
of discriminate
bricks,
and discriminate all them.
TakingTaking
chemical
them. el-
into
into consideration
consideration previousprevious
studies studies
related related
to the to the statistical
statistical analysis
analysis of of bricks,
bricks, all all chemical
chemical el-
ements were used in the analysis as they all may influence the discrimination and need to
elements
ements werewere
usedused in the
in the analysis
analysis as they
as they all may
all may influence
influence thethe discrimination
discrimination andand need
need to to
be considered [30,31].
be As can [30,31].
considered be observed
As can in observed
be Figure 6,inthe PCA
Figure 6, helped
the PCA distinguish
helped mainly
distinguish mainly
be considered [30,31]. As can be observed in Figure 6, the PCA helped distinguish mainly
two types of bricks
two
two and
types mortars.
types
of of bricks
bricks andand mortars.
mortars.

Figure 6. PCA of the bricks (c) and mortars (m).

Figure
Figure 6. PCA of the 6. PCA
bricks of the
(c) and bricks (c)(m).
mortars and mortars (m).
Buildings 2023, 13, x FOR PEER REVIEW 7 of 12
Buildings 2023,
Buildings 13,13,
2023, 321x FOR PEER REVIEW 7 7ofof12
11

3.3.
3.3.Classification
3.3.Classification
Classification
The
The
Theselection
selection
selection of the
ofofthe endmembers
endmembers
the endmembers was
waswas the
the most
most
the most important
important
important step
step iningenerating
step ingenerating
generating the
the re-
results
the re-
sults of
ofsults the supervised
the supervised classification.
classification.
of the supervised Based
Based on
classification. on the preliminary
the preliminary
Based imagistic
imagisticimagistic
on the preliminary analyses
analyses analyses corre-
correlated with
corre-
lated
the with
PCA
lated withthethe
PCA
results, results,
seven
PCA seven
endmembers
results, sevenendmembers were
were extracted
endmembers were extracted
extractedaccording
according toaccording totothe
thecharacter-
the characteristic spectral
character-
istic spectral
profiles profiles
of the materials:
istic spectral of the
profiles ofthree materials: three
types of bricks
the materials: types of
(#1, #2,
three types bricks
ofand (#1, #2,
#3),(#1,
bricks three and
#2,types #3), three
threetypes
of mortars
and #3), (#1,of#2,
types of
mortars
and
mortars (#1,
#3), and #2,
(#1,oneand #3),
material
#2, and and one
thatone
#3), and material
visually that
seemed
material visually seemed
impregnated
that visually seemed impregnated
in the surface. As
impregnated in the surface.
inpresented
the surface. in
As presented
Figure
As 7, red in
presented Figure
markers
in Figure 7, 7,
werered markers
attributed
red markerswere
to brick
were attributed
#1, greentoto
attributed tobrick
brick#1,
brick #1,green
#2, blue
green toto
to brick
brick
brick #2,
#3,
#2, blue
yellow
blue
tototo
brick
mortar
brick#3,#3,
yellow
#1, cyanto
yellow mortar
totomortar
mortar #1,
#2,cyan
#1,and to mortar
magenta
cyan to mortar #2,#2,
to and
mortar
andmagenta
magentatotomortar
#3. mortar#3. #3.

Figure
Figure7.7.ROI
Figure selection
7.ROI
ROI ofofof
selection
selection endmembers on
endmembers
endmembers the
on
on FCIR
the
the image.
FCIR
FCIR image.
image.

For
For
For LSU,
LSU,
LSU, wewe
we used
used
usedaaweighted
aweighted
weighted unit-sum
unit-sumconstraint,
unit-sum constraint,sosothe
constraint, theresults
resultswere
wereininthe
therange
the range
range
from
from 00 toto1.
1. For
For each
each endmember
endmember selected,
selected,the
the algorithm
algorithm
from 0 to 1. For each endmember selected, the algorithm generated generated
generated a specific
a specificgreyscale
greyscale
greyscale
image
image
image ininin
which
which
which light
light tones
lighttonesdesignate
tonesdesignate
designate higher
higherprobabilities
higher probabilities(a(abetter
probabilities betterfitting),
fitting),while
whiledarker
while darker
darker
tones
tones relate
tones relate
relatetototo
smaller
smaller
smaller probabilities
probabilities(see
probabilities (seeFigure
(seeFigure8).
Figure 8).
8).

(a)(a) (b)
(b)

(c)(c) (d)
(d)

(e)(e) (f)(f)
Figure
Figure
Figure 8.LSU
8.8.LSULSU classification
classification
classification results
results forfor
for (a)(a)
(a) brick
brick
brick #1,#1,
#1, (b)(b)
(b) brick
brick
brick #2,#2,
#2, (c) brick
(c)(c)brick
brick #3,(d)
#3,#3,
(d) (d)mortar
mortar
mortar #1,(e)
#1,#1,
(e) (e)mor-
mor-
mortar #2,
tar#2,
tar#2, andand(f) (f) mortar
mortar #3.
#3.
and (f) mortar #3.

The LSU classification depicts the adherent deposit’s presence on more areas than
were initially identified in the initial HSI analysis, as can be seen in Figure 9
The SAM classification was applied, and the threshold for each category of class was
0.100 radians. The number of unclassified pixels was insignificant, and the mapping of the
generated classes has great accuracy.
Figure 10 presents the results obtained for one of the seven acquisitions made in
order to cover the interest area. It was very interesting to highlight the distribution of
materials on the surface: the red hue is used for depicting the areas where brick #1 had a
higher probability, the green hue for brick #2, the blue hue for brick #3, the yellow hue for
Buildings 2023, 13, 321 8 of 11

mortar #1, the cyan hue for mortar #2, the magenta hue for mortar #3, and the kaki hue for
the impregnated layer. Error! Refer
not found.Figure 9. LSU classification for the adherent deposit layer.

The SAM classification was applied, and the threshold for each category o
0.100 radians. The number of unclassified pixels was insignificant, and the map
generated classes has great accuracy.
Figure 10 presents the results obtained for one of the seven acquisitions m
der to cover the interest area. It was very interesting to highlight the distributio
rials on the surface: the red hue is used for depicting the areas where brick #1 ha
probability, the green hue for brick #2, the blue hue for brick #3, the yellow hue
#1, the cyan hue for mortar #2, the magenta hue for mortar #3, and the kaki h
impregnated
Figure 9. LSU layer.
Figure 9. LSU classification for the adherent deposit layer.
classification for the adherent deposit layer.

The SAM classification was applied, and the threshold for each category of
0.100 radians. The number of unclassified pixels was insignificant, and the mapp
generated classes has great accuracy.
(a) obtained for one of the seven acquisitions ma
Figure 10 presents the results
der to cover the interest area. It was very interesting to highlight the distribution
rials on the surface: the red hue is used for depicting the areas where brick #1 had
probability, the green hue for brick #2, the blue hue for brick #3, the yellow hue f
(b)
#1, the cyan hue for mortar #2, the magenta hue for mortar #3, and the kaki hu
impregnated layer.

(c)

(a)

(d)

(b)

(e)

(c)

(f)

(d)

(g)
Figure 10.SAM
Figure 10. SAM classification
classification results forresults
(e)for(b)(a)
(a) brick #1, brick
brick #2, (c)#1,
brick(b)
#3, brick #2,#1,(c)
(d) mortar (e) brick #3, (d) m
mortar #2,
mortar #2,
(f) mortar #3,(f)
andmortar #3, and (g) impregnated.
(g) impregnated.

Figure 11 presents the final image obtained by overlapping all the SAM
tions (red, green, blue, yellow, cyan, magenta, and kaki) for the whole investiga
all seven acquisitions stitched together in one image.
 (g)
Buildings 2023, 13, 321 9 of 11 #1, (e)
Figure 10. SAM classification results for (a) brick #1, (b) brick #2, (c) brick #3, (d) mortar
mortar #2, (f) mortar #3, and (g) impregnated.

Figure1111
Figure presents
presents the the
finalfinal
image image obtained
obtained by overlapping
by overlapping all the
all the SAM SAM classifica-
classifications
(red,
tionsgreen, blue, yellow,
(red, green, cyan, magenta,
blue, yellow, and kaki)
cyan, magenta, forkaki)
and the whole
for theinvestigated area—allarea—
whole investigated
seven acquisitions stitched together in one image.
all seven acquisitions stitched together in one image.

Figure11.
Figure 11.Spectral
Spectral angle
angle mapper
mapper (SAM)
(SAM) classification.
classification.

Using
UsingPCA,PCA,SAM,SAM, and and LSULSU wewe cancandiscriminate
discriminate brick#1 fromfrom
brick#1 brick brick
#2 and#2#3and and#3 and
mortar #1 from mortar #2 and #3. Taking into consideration the fact
mortar #1 from mortar #2 and #3. Taking into consideration the fact that the long brick, that the long brick,
known
knowntotobebea recent
a recentmaterial,
material, is classified as brick
is classified #1, we#1,
as brick can
weseecanthesee
extent
the of the restora-
extent of the resto-
tion. The LIBS PCA data can clearly differentiate brick #1 from the other two, but the
ration. The LIBS PCA data can clearly differentiate brick #1 from the other two, but the
LIBS-HSI corroboration was able to depict a slight difference between brick #2 and #3. The
LIBS-HSI corroboration was able to depict a slight difference between brick #2 and #3. The
LIBS analyses did not identify Fe in the grey mortar areas attributed to mortar #1, and
LIBS
the analyses
data did not to
corresponding identify
it in theFe PCAin the
wasgrey mortar
grouped areas
apart fromattributed
all of them;to mortar
thus, we#1, and the
can
assume that mortar #1 is a new material that does not contain brick shards. We can observewe can
data corresponding to it in the PCA was grouped apart from all of them; thus,
assume
in Figurethat mortar
11 that it was#1used
is a new material that
for restoration does areas
on more not contain bricktoshards.
in addition the ones Wearound
can observe
in Figure
the 11 thatbricks.
new inserted it wasMortar
used for #2 restoration
and #3 contain on more areas in
brick shards andaddition
tend toto the ones
group in thearound
the new
PCA. inserted
In Figure bricks.
11e,f, we can Mortar
observe #2that
andthey
#3 contain brick shards
have a similar response; and tend to we
therefore, groupcan in the
assume
PCA. Inthat they 11e–f,
Figure are thewe same canmaterial,
observeand thatthe difference
they have a is made by
similar the thickness
response; of thewe can
therefore,
surface
assumeadherent
that theydeposit
are thelayer.
same It is interesting
material, andtothe
seedifference
also the presence
is made in by
traces
theof mortars of the
thickness
on these bricks—traces that are not visible with the naked eye. In addition,
surface adherent deposit layer. It is interesting to see also the presence in traces of mortars the presence
of
onsome
theseimpregnated
bricks—tracesmaterial
that isarenoted, a material
not visible withwith
theanaked
more distinctive spectral the
eye. In addition, profile
presence
than the rest, that will be the subject of further studies in order to see if it can be correlated
of some impregnated material is noted, a material with a more distinctive spectral profile
with the early development of Algae or Lichens [32], and deep learning algorithms are
than the rest, that will be the subject of further studies in order to see if it can be correlated
considered to be applied [33].
withThetheresults
earlyofdevelopment of Algae
the LSU were similar or SAM
to the Lichens [32], and meaning
classification, deep learning
that thealgorithms
selection are
considered to be applied [33].
of the endmembers was accurate. However, the root mean square (RMS) fraction displayed
Buildings 2023, 13, x FOR PEER REVIEW
some The
errorsresults of the LSUinsignificant),
(proportionally were similar to the SAM
depicted classification,
in Figure 12, thatmeaning
correlatedthat the selec-
to the
tion data,
LIBS of thedisplaying
endmembers a high was accurate.
content of C However,
which can be theassociated
root meantosquare
smoke (RMS)scraps onfraction
the dis-
played
outer some errors (proportionally insignificant), depicted in Figure 12, that correlated to
walls.
the LIBS data, displaying a high content of C which can be associated to smoke scraps on
the outer walls.

Figure 12. SAM error.

Figure 12. SAM error.

4. Conclusions
This paper discusses the results obtained by corroborating LIBS an
Buildings 2023, 13, 321 10 of 11

4. Conclusions
This paper discusses the results obtained by corroborating LIBS and HSI data acquired
on an area of the Turkish bath from the Goles, ti Museum, which was selected because it is
representative in terms of age value, covering both original and restoration materials. The
two complementary techniques were used in order to map the original elements and trace
the undocumented interventions previously made on the Turkish bath. Using the FCIR
mode, RGB channels were assigned to different combinations of wavelengths ranging from
954 to 2250 nm in order to track the similarities and differences in the material behaviour,
and a total of seven areas were selected for endmembers’ attribution. LIBS stratigraphy was
performed based on the HSI results. The main chemical elements identified using LIBS can
be traced to oxides commonly found in clay bricks, and the PCA helped distinguish two
main types of bricks and mortars. The LIBS PCA results were correlated and, further on,
processed in ENVI using SAM and LSU, applied complementary, in order to validate the
accuracy of the classification and mapping, as seven endmembers were extracted for three
brick areas (#1, #2, and #3), three mortar areas (#1, #2, and #3), and one material considered
to be impregnated in the surface. Using PCA, SAM, and LSU, a distribution map of the area
was obtained where previous interventions were identified and mapped. The impregnated
material detected will be the subject of further studies in order to see if it is a biological
attack. Based on the results, deep learning algorithms are considered to be used for further
applications on the building.

Author Contributions: Conceptualization, M.D.; historical input, C.C. and G.C.; methodology, M.D.;
HIS acquisition and processing, L.C.R.; LIBS acquisition and processing, M.D.; writing—original
draft preparation, M.D.; writing—review and editing, M.D., L.C.R., C.C. and G.C. All authors have
read and agreed to the published version of the manuscript.
Funding: This research was funded by the Romanian Ministry of Research, Innovation and Digitaliza-
tion, under Program 1—Development of the National Research-Development System, Subprogram
1.2—Institutional performance—Projects to finance the excellence in RDI, SUPECONEX grant nr.
18PFE/30.12.2021, under Core Program 18N/08.02.2019, project PN 19-18.01.02 and under PNCDI III,
CNCS—UEFISCDI, project number PN-III-P4-PCE-2021-1605.
Data Availability Statement: The data that support the findings of this study are available upon
request from the corresponding authors.
Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design
of this study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or
in the decision to publish the results.

References
1. Fernandes, F.M.; Lourenço, P.B.; Castro, F. Ancient Clay Bricks: Manufacture and Properties. In Materials, Technologies and Practice
in Historic Heritage Structures; Springer: Berlin/Heidelberg, Germany, 2010. [CrossRef]
2. Teodoru, R. Curt, i întărite târzii. Stud. S, i Cercet. Istor. Artei 1963, X, 343.
3. Din Alep, P. Jurnal de călătorie în Moldova s, i Valahia; Ioana Feodorov editions; Romanian Academy Publishing House and Istros
Publishing House: Bucharest, Romania, 2014.
4. Potra, G. Din Bucures, tii de Altădată; Scientific and Encyclopedic Publishing House: Bucharest, Romania, 1981.
5. Galbács, G. A critical review of recent progress in analytical laser-induced breakdown spectroscopy. Anal. Bioanal. Chem. 2015,
407, 7537–7562. [CrossRef] [PubMed]
6. Carmona-Quiroga, P.M.; Martínez-Ramírez, S.; Sánchez-Cortés, S.; Oujja, M.; Castillejo, M.; Blanco-Varela, M.T. Effectiveness of
antigraffiti treatments in connection with penetration depth determined by different techniques. J. Cult. Herit. 2010, 11, 297–303.
[CrossRef]
7. Palleschi, V. (Ed.) Chemometrics and Numerical Methods in LIBS. Wiley: Hoboken, NJ, USA, 2022.
8. Zahiri, Z.; Laefer, D.F.; Kurz, T.; Buckley, S.; Gowen, A. A comparison of ground-based hyperspectral imaging and red-edge
multispectral imaging for façade material classification. Autom. Constr. 2022, 136, 104164. [CrossRef]
9. Popescu, M.; Iliescu, C. Goles, ti; Meridiane Publishing House: Bucharest, Romania, 1966.
10. Pagnotta, S.; Lezzerini, M.; Campanella, B.; Gallello, G.; Grifoni, E.; Legnaioli, S.; Lorenzetti, G.; Poggialini, F.; Raneri, S.; Safi, A.;
et al. Fast quantitative elemental mapping of highly inhomogeneous materials by micro-Laser-Induced Breakdown Spectroscopy.
Spectrochim. Acta Part B At. Spectrosc. 2018, 146, 9–15. [CrossRef]
Buildings 2023, 13, 321 11 of 11

11. Živković, S.; Botto, A.; Campanella, B.; Lezzerini, M.; Momčilović, M.; Pagnotta, S.; Palleschi, V.; Poggialini, F.; Legnaioli, S.
Laser-Induced Breakdown Spectroscopy elemental mapping of the construction material from the Smederevo Fortress (Republic
of Serbia). Spectrochim. Acta Part B At. Spectrosc. 2021, 181, 106219. [CrossRef]
12. Dontu, S.; Miclos, S.; Savastru, D.; Tautan, M. Combined spectral-domain optical coherence tomography and hyperspectral
imaging applied for tissue analysis: Preliminary results. Appl. Surf. Sci. 2017, 417, 119–123. [CrossRef]
13. Cheng, J.H.; Nicolai, B.; Sun, D.W. Hyperspectral imaging with multivariate analysis for technological parameters prediction and
classification of muscle foods: A review. Meat Sci. 2017, 123, 182–191. [CrossRef]
14. Ravikanth, L.; Jayas, D.S.; White, N.D.G.; Fields, P.G.; Sun, D.W. Extraction of Spectral Information from Hyperspectral Data and
Application of Hyperspectral Imaging for Food and Agricultural Products. Food Bioprocess Technol. 2017, 10, 1–33. [CrossRef]
15. Spizzichino, V.; Fantoni, R. Laser Induced Breakdown Spectroscopy in archeometry: A review of its application and future
perspectives. Spectrochim. Acta Part B At. Spectrosc. 2014, 99, 201–209. [CrossRef]
16. Gaudiuso, R.; Dell’Aglio, M.; de Pascale, O.; Senesi, G.S.; de Giacomo, A. Laser induced breakdown spectroscopy for elemental
analysis in environmental, cultural heritage and space applications: A review of methods and results. Sensors 2010, 10, 7434–7468.
[CrossRef] [PubMed]
17. Hahn, D.W.; Omenetto, N. Laser-induced breakdown spectroscopy (LIBS), part II: Review of instrumental and methodological
approaches to material analysis and applications to different fields. Appl. Spectrosc. 2012, 66, 347–419. [CrossRef]
18. Botto, A.; Campanella, B.; Legnaioli, S.; Lezzerini, M.; Lorenzetti, G.; Pagnotta, S.; Poggialini, F.; Palleschi, V. Applications of
laser-induced breakdown spectroscopy in cultural heritage and archaeology: A critical review. J. Anal. At. Spectrom. 2019, 34,
81–103. [CrossRef]
19. Atanassova, V.; Ghervase, L.; Cortea, I.M.; Mihailov, V.; Tankova, V.; Nikolov, V. Multi-analytical approach for characterization
of archaeological pottery excavated in the Early-Neolithic settlement of Chavdar, Bulgaria. Spectrosc. Lett. 2021, 54, 549–559.
[CrossRef]
20. Pacher, U.; Dinu, M.; Nagy, T.O.; Radvan, R.; Kautek, W. Multiple wavelength stratigraphy by laser-induced breakdown
spectroscopy of Ni-Co alloy coatings on steel. Spectrochim. Acta Part B At. Spectrosc. 2018, 146, 36–40. [CrossRef]
21. Ghervase, L.; Dinu, M.; Bors, , C.; Anghelut, ă, L.M.; Rădvan, R.; Cortea, I.M. Investigation on Metal Adornments From Ancient
Eastern Europe. Front. Mater. 2020, 7, 600913. [CrossRef]
22. Cortea, I.M.; Ratoiu, L.; Ghervase, L.; T, entea, O.; Dinu, M. Investigation of ancient wall painting fragments discovered in the
roman baths from alburnus maior by complementary non-destructive techniques. Appl. Sci. 2021, 11, 10049. [CrossRef]
23. Cucci, C.; Picollo, M.; Chiarantini, L.; Uda, G.; Fiori, L.; De Nigris, B.; Osanna, M. Remote-sensing hyperspectral imaging for
applications in archaeological areas: Non-invasive investigations on wall paintings and on mural inscriptions in the Pompeii site.
Microchem. J. 2020, 158, 105082. [CrossRef]
24. Miliani, C.; Rosi, F.; Daveri, A.; Brunetti, B.G. Reflection infrared spectroscopy for the non-invasive in situ study of artists’
pigments. Appl. Phys. A Mater. Sci. Process. 2012, 106, 295–307. [CrossRef]
25. Sánchez, J.; Quirós, E. Semiautomatic detection and classification of materials in historic buildings with low-cost photogrammetric
equipment. J. Cult. Herit. 2017, 25, 21–30. [CrossRef]
26. Canty, M.J. Image Analysis, Classification and Change Detection in Remote Sensing: With Algorithms for ENVI/IDL and Python, 3rd ed.;
CRC Press: Boca Raton, FL, USA, 2014; ISBN 9781466570382.
27. Bai, D.; Messinger, D.W.; Howell, D. A hyperspectral imaging spectral unmixing and classification approach to pigment mapping
in the Gough & Selden Maps. J. Am. Inst. Conserv. 2019, 58, 69–89. [CrossRef]
28. Allios, D.; Guermeur, N.; Cocoual, A.; Linderholm, J.; Sciuto, C.; Geladi, P.; Gobrecht, A.; Bendoula, R.; Moura, D.; Jay, S.; et al.
Near Infrared Spectra and Hyperspectral Imaging of Medieval Fortress Walls in Carcassonne: A Comprehensive Interdisciplinary
Field Study. NIR News 2016, 27, 16–20. [CrossRef]
29. Bernstein, L.S. Quick atmospheric correction code: Algorithm description and recent upgrades. Opt. Eng. 2012, 51, 111719.
[CrossRef]
30. Capedri, S.; Venturelli, G. Provenance determination of trachyctic lavas, employed as blocks in the Romanesque cathedral of
Modena (Northern Italy), using magnetic susceptibility, and petrographic and chemical parameters. J. Cult. Herit. 2005, 6, 7–19.
[CrossRef]
31. Gomoiu, R.I.; Cojoc, R.; Ruginescu, S.; Neagu, M.; Enache, G.; Maria, M.; Dumbrăvician, I.; Olteanu, R.; Rădvan, L.; Ratoiu, V.; et al.
Brackish and Hypersaline Lakes as Potential Reservoir for Enzymes Involved in Decomposition of Organic Materials on Frescoes.
Fermentation 2022, 8, 462. [CrossRef]
32. Angheluta, L.; Striber, J.; Radvan, R.; Gomoiu, I.; Dragomir, V. Non-contact and non-invasive photonic device for qualitative
fungal contamination control. Math. Comput. Biol. Chem. 2008, 4, 147–156.
33. Angheluţă, L.M.; Chiroşca, A. Physical degradation detection on artwork surface polychromies using deep learning models.
Rom. Rep. Phys. 2020, 72, 805.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.