Documenting Paintings with Gigapixel Photography - MDPI

Journal of
 Imaging

Article
Documenting Paintings with Gigapixel Photography
Pedro M. Cabezos-Bernal 1, * , Pablo Rodriguez-Navarro 2 and Teresa Gil-Piqueras 2

 1 Departamento de Expresión Gráfica Arquitectónica, Universitat Politècnica de València, 46022 Valencia, Spain
 2 Centro de Investigación en Arquitectura, Patrimonio y Gestión para el Desarrollo Sostenible (PEGASO),
 Universitat Politècnica de València, 46022 Valencia, Spain; rodriguez@upv.es (P.R.-N.);
 tgil@ega.upv.es (T.G.-P.)
 * Correspondence: pcabezos@ega.upv.es

 Abstract: Digital photographic capture of pictorial artworks with gigapixel resolution (around 1000
 megapixels or greater) is a novel technique that is beginning to be used by some important interna-
 tional museums as a means of documentation, analysis, and dissemination of their masterpieces.
 This line of research is extremely interesting, not only for art curators and scholars but also for the
 general public. The results can be disseminated through online virtual museum displays, offering
 a detailed interactive visualization. These virtual visualizations allow the viewer to delve into the
 artwork in such a way that it is possible to zoom in and observe those details, which would be
 negligible to the naked eye in a real visit. Therefore, this kind of virtual visualization using gigapixel
 images has become an essential tool to enhance cultural heritage and to make it accessible to every-
 one. Since today’s professional digital cameras provide images of around 40 megapixels, obtaining
 gigapixel images requires some special capture and editing techniques. This article describes a series
 



 
 of photographic methodologies and equipment, developed by the team of researchers, that have been
 put into practice to achieve a very high level of detail and chromatic fidelity, in the documentation
Citation: Cabezos-Bernal, P.M.;
 and dissemination of pictorial artworks. The result of this research work consisted in the gigapixel
Rodriguez-Navarro, P.; Gil-Piqueras,
 documentation of several masterpieces of the Museo de Bellas Artes of Valencia, one of the main
T. Documenting Paintings with
Gigapixel Photography. J. Imaging
 art galleries in Spain. The results will be disseminated through the Internet, as will be shown with
2021, 7, 156. http://doi.org/10.3390/ some examples.
jimaging7080156
 Keywords: gigapixel photography; ultra-high resolution; art documentation; stitching; image regis-
Academic Editors: Guillaume Caron, tration; virtual musealization
Olga Regina Pereira Bellon and
Ilan Shimshoni

Received: 14 July 2021 1. Introduction
Accepted: 18 August 2021
 New techniques for the dissemination of cultural heritage through digital media
Published: 21 August 2021
 is one of the lines of research providing benefits to society since the research results
 can be available to the public by means online virtual museum displays. In fact, many
Publisher’s Note: MDPI stays neutral
 important international museums [1], are using these technologies in collaboration with
with regard to jurisdictional claims in
 the multinational Google, which has developed its own high-resolution digital capture
published maps and institutional affil-
 system for paintings. These works are exhibited on the Arts and Culture Project website [2],
iations.
 which contains many artworks that can be displayed with a high level of detail. They are
 digital reproductions with a gigapixel resolution, that is, a resolution greater than 1000
 megapixels, which is 50 times greater than the image resolution provided by a conventional
 digital camera.
Copyright: © 2021 by the authors.
 Gigapixel images allow documenting and analysing the paintings accurately, which is
Licensee MDPI, Basel, Switzerland.
 very useful for curators and art scholars. Furthermore, the virtual visualizations that can
This article is an open access article
 be generated from this type of image make the artwork accessible to anyone connected to
distributed under the terms and
 the Internet. The viewers will be able to dive into the work, in such a way, that they would
conditions of the Creative Commons
 appreciate many details, which would be negligible to the eye in a real visit (Figure 1).
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).

J. Imaging 2021, 7, 156. https://doi.org/10.3390/jimaging7080156 https://www.mdpi.com/journal/jimaging

J. Imaging 2021, 7, 156 2 of 20
 J. Imaging 2021, 7, x FOR PEER REVIEW 2 of 20

 Figure 1.1.Hans
 Figure HansHolbein
 Holbein
 thethe Younger,
 Younger, 1533,
 1533, The Ambasssadors.
 The Ambasssadors. Picture
 Picture and obtained
 and detail detail obtained from
 from Google
 Google Arts and Culture
 Arts and Culture [2]. [2].

 Apart from
 Apart from Google,
 Google, therethere are very very few companies specialized in capturing gigapixel gigapixel
 images of artworks
 images of artworks due to due to the technical complexity and the specialized equipment. There There
 are some
 are someexamples,
 examples,such suchas asthe
 theFrench
 Frenchstate stateorganization
 organizationCentre CentrededeRecherche
 RechercheetetdedeRestaura-
 Restau-
 ration
 tion desdes Musées
 Musées de de France
 France CR2MF
 CR2MF [3],[3],
 thethe Italian
 Italian company
 company Haltadefinizione
 Haltadefinizione [4],[4],
 ororthethe Span-
 Spanish
 ish Madpixel
 Madpixel [5]. Some
 [5]. Some researchers
 researchers have have used usedgigapixelgigapixel photography
 photography for documenting
 for documenting rupes-
 rupestrian
 trian paintingspaintings
 [6] and[6]evenandcombined
 even combined with multispectral
 with multispectral imaging imaging
 techniques techniques [7].
 [7]. Other
 Other research works have used gigapixel images for
 research works have used gigapixel images for documenting large Baroque illusionistic documenting large Baroque illu-
 sionistic paintings
 paintings [8,9]. [8,9].
 Digital photographic
 Digital photographic capture capture withwith gigapixel
 gigapixel resolution
 resolution is is not an easy task and and serious
 serious
 difficulties may arise
 difficulties arisedue duetotophysical
 physicalproblems problems such
 suchas the
 as thediffraction
 diffractionof light, which
 of light, rep-
 which
 resents a barrier
 represents a barrierthatthat
 limits the the
 limits sharpness
 sharpness thatthat
 can canbe obtained
 be obtained withwith
 an optical
 an opticaldevice and
 device
 a digital
 and sensor.
 a digital For this
 sensor. reason,
 For this the progression
 reason, the progression in the in resolution
 the resolution of digital sensors
 of digital has
 sensors
 already
 has reached
 already reachedthe limit established
 the limit established by theby diffraction of lightofand
 the diffraction byand
 light the optical
 by the resolu-
 optical
 tion provided
 resolution by theby
 provided lenses [10] that
 the lenses [10] cannot match match
 that cannot the resolution of the current
 the resolution top digital
 of the current top
 digital
 sensors. sensors. Going beyond
 Going beyond this would this would
 not take not take advantage
 advantage of the effective
 of the effective resolution resolution
 of the
 of the sensor,
 sensor, unlessunless
 the sizethe ofsize
 the of the optical-sensor
 optical-sensor assembly assembly were increased,
 were increased, which is which is not
 not practi-
 practical
 cal for the forfuture
 the future development
 development of conventional
 of conventional cameras.
 cameras. In fact,
 In fact, therethere
 havehave beenbeenfewfew at-
 attempts
 tempts totodevelop
 developprototype
 prototypecameras camerasprovidingprovidinggigapixel
 gigapixelimagesimageson onthe thefly.
 fly. One
 One of of the
 the
 first
 first was
 was the large format format camera
 cameradeveloped
 developedwithin withinthe Gigapxl
 theGigapxl Project
 Project byby Graham
 Graham Flint
 Flint in
 in 2000 [11]. In this bulky camera, the image was exposed
 2000 [11]. In this bulky camera, the image was exposed on a 450 × 225 mm negative film, on a 450 × 225 mm negative
 film,
 which which was digitized
 was later later digitized
 to formtoa form a 4 gigapixel
 4 gigapixel (Gp) image. (Gp)Another
 image. Another
 more recent more recent
 proposal
 proposal
 is the AWARE is the 2AWARE
 gigapixel 2 gigapixel
 camera, started camera, by started
 D. Brady byandD. his
 Bradyteam and his team
 in 2012. Thisinproto-
 2012.
 This
 type prototype
 included 98 included 98 micro-cameras
 micro-cameras with a combined with a combined
 resolution resolution
 of 1 Gp and ofit1 isGp and
 still it is
 being
 still being developed
 developed and improved. and improved.
 The AWARE The10AWARE 10 and reach
 and 40 models 40 modelsalmost reach
 3 Gp. almost 3 Gp.
 Neverthe-
 Nevertheless,
 less, such special such special prototypes
 prototypes are far from arebecoming
 far from becoming
 an accessible an accessible
 option dueoption to theirduehigh to
 their high complexity
 complexity and cost. and cost.
 A
 A reasonable
 reasonable solution
 solution to overcome the problem of diffraction and achieve gigapixel gigapixel
 images using conventional
 images using conventional cameras is multi-shot panoramic capture, which consists
 capture, which consists in in
 obtaining
 obtaining aa set set of
 of photographs
 photographs from from the the same
 same point
 point of of view
 view andand with
 with aa sufficient
 sufficient overlap
 overlap
 between
 betweenadjacent
 adjacentpictures
 picturesso so that,
 that, by
 by means
 means of image stitching
 of image stitching software,
 software, they they cancan bebe joined
 joined
 to
 to compose
 compose aa higher
 higher resolution
 resolution image image [12]. [12].
 In
 In order
 order to to achieve
 achieve aa perfect
 perfect stitch
 stitch between
 between pictures
 pictures andand to to avoid
 avoid parallax
 parallax errors,
 errors, it it is
 is
 mandatory to use a panoramic head (Figure 2) to fix the
 mandatory to use a panoramic head (Figure 2) to fix the position of the optical centre orposition of the optical centre or
 no-parallax
 no-parallax pointpoint of of the
 the lens,
 lens, while
 while rotating
 rotating the the camera
 camera to to obtain
 obtain thethe different
 different shotsshots that
 that
 will compose the final image. Moreover, a telephoto lens should be used so as to maximize
 the final resolution.

J. Imaging 2021, 7, x FOR PEER REVIEW 3 of 20

J. Imaging 2021, 7, 156 3 of 20
 will compose the final image. Moreover, a telephoto lens should be used so as to maximize
 the final resolution.

 Figure 2.
 Figure 2. Manual
 Manual and
 and motorized
 motorized panoramic
 panoramic heads.
 heads.

 The capture system developed developed by by Google,
 Google, called
 called Art Art Camera
 Camera[13], [13],seem
 seemto tobe bebased
 based
 on this
 this same
 sameprinciple,
 principle,and andconsists
 consistsof of
 a camera
 a cameraintegrated
 integrated in a in
 panoramic
 a panoramic head headthat pro-
 that
 gressively sweeps the painting from a fixed point. The partial images
 progressively sweeps the painting from a fixed point. The partial images are subsequently are subsequently
 processed by the multinational itself. This type type of
 of camera
 camera is is not
 notfor
 forsale
 saleand
 andthere
 thereareareonly
 only
 a very limited number of units available exclusively exclusively to to the
 the company.
 company.
 The capture systems
 systems used used by
 by other
 other commercial
 commercialcompanies
 companiessuch suchas asHaltadefinizione,
 Haltadefinizione,
 not fully
 are not fully described,
 described,but butsome
 someofofthem
 themarearebased
 basedonon using
 using high-end
 high-end digital
 digital reflex
 reflex cam-
 cameras
 eras high-cost
 and and high-cost automated
 automated panoramic
 panoramic headsheads by Clauss
 by Clauss [14]. Madpixel
 [14]. Madpixel has hishas ownhisautomatic
 own au-
 panoramic head called
 tomatic panoramic head MadpixelROB [15]. The[15].
 called MadpixelROB aforementioned research works
 The aforementioned research have also
 works
 used the panoramic head methodology.
 have also used the panoramic head methodology.
 Using aa panoramic
 panoramichead headandanda atelephoto
 telephoto lens
 lensis aisvery
 a veryeffective method
 effective methodfor document-
 for docu-
 ing relatively
 menting smallsmall
 relatively paintings withwith
 paintings gigapixel resolution,
 gigapixel resolution, butbutwhenwhen it comes
 it comes to to
 capturing
 captur-
 artworks
 ing artworks of moderate
 of moderate size, there
 size, arise
 there some
 arise drawbacks
 some drawbacks that limit
 that image
 limit image sharpness
 sharpness due
 dueto
 narrow
 to narrow depth
 depthof field provided
 of field by long
 provided focalfocal
 by long length lenses.
 length This This
 lenses. problem will be
 problem discussed
 will be dis-
 and
 cussedfixed
 andbyfixed
 following the techniques
 by following that will
 the techniques thatbewill
 revealed below,below,
 be revealed since since
 one ofone theofmain
 the
 aims of this article is to show a new and very accurate gigapixel capture
 main aims of this article is to show a new and very accurate gigapixel capture methodol- methodology that
 uses relatively
 ogy that affordable
 uses relatively equipment.
 affordable Moreover,
 equipment. the digital
 Moreover, theimage
 digitalprocessing to generate
 image processing to
 the gigapixel
 generate image and
 the gigapixel the dissemination
 image of the results
 and the dissemination onresults
 of the the Internet
 on thewill be explained
 Internet will be
 and carriedand
 explained outcarried
 by means out ofbyopensource software. software.
 means of opensource

 2. Materials and
 2. Materials and Methods
 Methods
 2.1. Capture Techniques
 2.1. Capture Techniques
 The photographic equipment that was used in this research work consisted of a
 The photographic equipment that was used in this research work consisted of a 32.5
 32.5 Mp digital camera Canon EOS 90D, equipped with a fixed telephoto lens Canon EF
 Mp digital camera Canon EOS 90D, equipped with a fixed telephoto lens Canon EF 200
 200 mm f2.8 L II USM, a Manfrotto tripod 28B and a Manfrotto panoramic head 303 SPH.
 mm f2.8 L II USM, a Manfrotto tripod 28B and a Manfrotto panoramic head 303 SPH.
 Long focal length lenses should be used to maximize resolution and the camera must
 Long focal length lenses should be used to maximize resolution and the camera must
 be quite close to the artwork, so that, in these circumstances, depth of field provided by
 be quite close to the artwork, so that, in these circumstances, depth of field provided by
 the lens is very short. Furthermore, this is aggravated by the need to use an intermediate
 the lens is very short. Furthermore, this is aggravated by the need to use an intermediate
 diaphragm aperture to avoid losing global sharpness by the effect of light diffraction [16].
 diaphragm aperture to avoid losing global sharpness by the effect of light diffraction [16].
 This problem causes a loss of sharpness in certain zones of the off-centred shots of the
 This problem causes a loss of sharpness in certain zones of the off-centred shots of the
 paintings, when using the panoramic head capture methodology. This is due to the angular
 paintings, when using the panoramic head capture methodology. This is due to the angu-
 deviation between the camera sensor and the painting.
 lar deviation between the camera sensor and the painting.
 When the digital sensor remains parallel to the painting, all the points are perfectly
 When the digital sensor remains parallel to the painting, all the points are perfectly
 focused, so the distance between them and the sensor remain the same. However, as
 focused,
 the cameraso is
 therotated
 distance between
 towards thethem andmargins,
 picture the sensor
 theremain the same.the
 angle between However,
 sensor andas the
 the
 camera is rotated towards the picture margins, the angle between the sensor and
 painting increases dramatically (Figure 3). This causes only those points that are at thethe paint-
 ing increases
 same distancedramatically (Figure
 from the camera 3). This
 sensor causes
 as the only point
 focused those to
 points that are sharp,
 be perfectly at the same
 while
 those remaining at different distances, will be gradually blurred due to narrow depth of
 field (Figure 4).

J. Imaging 2021, 7, x FOR PEER REVIEW 4 of 20
 J. Imaging 2021, 7, x FOR PEER REVIEW 4 of 20

J. Imaging 2021, 7, 156 distance from the camera sensor as the focused point to be perfectly sharp, while those
 4 of 20
 distance from the camera sensor as the focused point to be perfectly sharp, while those
 remaining at different distances, will be gradually blurred due to narrow depth of field
 remaining at different distances, will be gradually blurred due to narrow depth of field
 (Figure 4).
 (Figure 4).

 Figure 3.
 Figure 3. Obliquity
 Obliquity between
 between the
 the shots
 shots and
 and the
 the canvas
 canvas when
 when using
 using aa panoramic
 panoramic head.
 head.
 Figure 3. Obliquity between the shots and the canvas when using a panoramic head.

 Figure 4. Loss of sharpness in an oblique shot due to narrow depth of field of long focal lenses.
 Figure 4.
 Figure 4. Loss
 Loss of
 of sharpness
 sharpness in
 in an
 an oblique
 oblique shot
 shot due
 due to
 to narrow
 narrow depth
 depth of
 of field
 field of
 of long
 long focal
 focal lenses.
 lenses.
 To solve this problem, a new capture technique has been put into practice, which
 To solve
 To solve this
 this problem,
 problem, aa new
 new capture
 capture technique
 technique has been put into practice, which which
 consists in taking the shots while moving the camera parallel to the painting (Figure 5).
 consists in
 consists in taking
 taking the shots while moving
 moving the
 the camera
 camera parallel
 parallel to
 tothe
 thepainting
 painting(Figure
 (Figure5).
 5).
 Doing so, the sharpness of the pictures will be always optimal and never limited by depth
 Doingso,
 Doing so,the
 thesharpness
 sharpnessofofthe
 thepictures
 pictureswill
 willbebealways
 alwaysoptimal
 optimal and
 and never
 never limited
 limited byby depth
 depth of
 of field. However, while this method would solve the problem of sharpness, it also gen-
 of field.
 field. However,
 However, whilewhile
 thisthis method
 method would
 would solvesolve the problem
 the problem of sharpness,
 of sharpness, it generates
 it also also gen-
 erates some difficulties that must be overcome. On the one hand, as the viewpoint would
 eratesdifficulties
 some some difficulties thatbe
 that must must be overcome.
 overcome. On theOn thehand,
 one one hand,
 as theasviewpoint
 the viewpoint
 would would
 vary
 vary continuously, the light reflected by the work will change between the shots, causing
 vary continuously,
 continuously, thereflected
 the light light reflected
 by theby the will
 work workchange
 will change
 betweenbetween the shots,
 the shots, causingcausing
 slight
 slight differences in exposure and problems with specular reflections. Those issues can be
 slight differences
 differences in exposure
 in exposure and problems
 and problems with specular
 with specular reflections.
 reflections. Those can
 Those issues issues
 be can be
 solved
 by using a strategically placed controlled light source, which moves along with the camera.

J. Imaging 2021, 7, x FOR PEER REVIEW 5 of 20

J. Imaging 2021, 7, 156 5 of 20
 solved by using a strategically placed controlled light source, which moves along with the
 camera.

 Figure 5.
 Figure 5. Parallel
 Parallel camera
 camera translation
 translation to
 to take
 take all
 all the
 the pictures
 pictures frontally
 frontally to
 to the
 the canvas.
 canvas.

 Another problem arises when it comes to joining the photo mosaic,
 Another mosaic, obtained
 obtained in in that
 that
 way, so
 way, so most of of the
 thecurrent
 currentsoftware
 softwareforfor stitching
 stitching images
 images would
 wouldbe unable
 be unableto join
 to the
 joinmo-
 the
 saic when
 mosaic when thetheviewpoint
 viewpoint is is
 not
 not static.
 static.Fortunately,
 Fortunately,this thisproblem
 problemcan canbebesolved
 solvedwithwith thethe
 help of
 help of some
 some stitching
 stitching algorithms,
 algorithms, initially developed by the German professor professor Helmut
 Helmut
 Dersch [17],
 Dersch [17],which
 whichwere wereimplemented
 implemented in in Hugin
 Hugin [18],
 [18], an opensource
 an opensource stitching
 stitching software
 software with
 with general
 general publicpublic
 licenselicense that allows
 that allows joiningjoining
 pictures pictures
 taken from takendifferent
 from different viewpoints,
 viewpoints, as long
 as they
 as longrepresent
 as they represent a planarOptionally,
 a planar surface. surface. Optionally,
 the commercial the commercial
 software PTGUI software
 [19]PTGUI
 would
 also
 [19] do an excellent
 would also do an jobexcellent
 when stitching
 job when these kinds these
 stitching of photomosaics.
 kinds of photomosaics.
 Another
 Another way wayto toconstruct
 constructthe thegigapixel
 gigapixelimage image is is
 byby means
 means SfMSfM(structure
 (structurefrom motion)
 from mo-
 photogrammetry
 tion) photogrammetry software, sincesince
 software, the construction
 the construction algorithm
 algorithm is conceived
 is conceived totogenerate
 generatea
 textured
 a textured mesh
 meshfrom frompictures
 picturestakentakenfrom fromdifferent
 differentviewpoints.
 viewpoints. The problem is that that itit
 would
 would be be needed
 needed aa greater
 greater number
 number of shots since a minimum overlap overlap of of 50%
 50% between
 between
 adjacent
 adjacent pictures is advisable, while a 30% overlap would would be fine when using using stitching
 stitching
 software.
 software. However, an advantage of using SfM photogrammetry software
 photogrammetry software would be would be the
 the
 3D
 3D reconstruction
 reconstruction of the painting and even the frame, which is not a plane plane surface,
 surface, while
 while
 stitching
 stitching software
 softwarewould wouldprovide
 provideananaccurateaccurate 2D2D rectified
 rectified orthophoto
 orthophoto of the canvas
 of the canvasbutbut
 not
 of
 notthe
 of volumetric
 the volumetric frame.frame.
 Taking
 Taking intointo account
 account the the previous
 previous considerations,
 considerations, the following gigapixel capture
 following gigapixel capture
 techniques
 techniques are are proposed,
 proposed, which which will be suitable for paintings or planar surfaces:
 The
 The single-viewpoint
 single-viewpointcapture capture technique
 technique is aisknown
 a known and common
 and common methodology,
 methodology,which
 consist in using a panoramic head to rotate the camera around
 which consist in using a panoramic head to rotate the camera around the no-parallax the no-parallax point, which
 would be fixed
 point, which and centred
 would be fixedin andfront of theincanvas,
 centred front ofthus obtaining
 the canvas, a set
 thus of overlapping
 obtaining a set of
 pictures
 overlapping(Figure 6). This
 pictures technique
 (Figure 6). Thisrequires
 techniquelimiting the obliquity
 requires limiting theof the shots inofrelation
 obliquity the shots to
 the canvas to
 in relation in the
 order to preserve
 canvas in order thetomaximum
 preserve the image
 maximumsharpness.image sharpness.
 The
 The parallel-multi-viewpoint
 parallel-multi-viewpoint capture capture technique
 technique is is aa new
 new method,
 method, whichwhich consist
 consist in in
 taking a mosaic of overlapped pictures as the camera is moved
 taking a mosaic of overlapped pictures as the camera is moved in parallel to the canvas in parallel to the canvas
 describing several rows
 describing several rowsor orcolumns.
 columns.The The camera
 camera digital
 digital sensor
 sensor must must be parallel
 be parallel to thetoart-
 the
 artwork, or, in other words, the lens optical axis should
 work, or, in other words, the lens optical axis should be orthogonal to the canvas (Figurebe orthogonal to the canvas
 (Figure
 7). 7).

J. Imaging 2021, 7, 156 6 of 20
 J. Imaging 2021, 7, x FOR PEER REVIEW 6 of 20

 J. Imaging 2021, 7, x FOR PEER REVIEW 6 of 20

 Figure6.
 Figure 6. Single-viewpoint
 Single-viewpoint capture
 capturetechnique.
 technique. The
 The camera
 camera isis rotated
 rotatedaround
 aroundits
 itsno-parallax
 no-parallaxpoint.
 point.
 Figure 6. Single-viewpoint capture technique. The camera is rotated around its no-parallax point.

 Figure 7. Parallel-multi-viewpoint capture technique. The camera moves parallel to the canvas.

J. Imaging 2021, 7, x FOR PEER REVIEW 7 of 20

J. Imaging 2021, 7, 156 7 of 20

 Figure 7. Parallel-multi-viewpoint capture technique. The camera moves parallel to the canvas.

 The
 The tilted-multi-viewpoint
 tilted-multi-viewpoint capturecapture technique
 technique isis aa variation
 variation ofof the
 the previous
 previous one,
 one, in
 in
 which
 which the camera can be tilted if required. This option may seem awkward but can be
 the camera can be tilted if required. This option may seem awkward but can be
 useful
 useful in
 in the
 the case
 case of
 of capturing
 capturing large
 largepaintings
 paintingswith
 withaatripod
 tripodthat
 thatisisnot
 nottall
 tallor
 or short
 shortenough
 enough
 to
 to allow the parallel movement of the camera in front of the whole painting. In
 allow the parallel movement of the camera in front of the whole painting. In that
 that case,
 case,
 the
 the camera
 camera could
 could be
 be tilted
 tilted when
 when taking
 taking the
 the upper
 upper or
 or lower
 lower rows
 rows to
 to cover
 cover the
 the entire
 entire canvas.
 canvas.
 (Figure
 (Figure 8).
 8).

 Figure 8.
 Figure 8. Tilted-multi-viewpoint
 Tilted-multi-viewpointcapture
 capturetechnique.
 technique.The camera
 The camerais moved mostly
 is moved parallel
 mostly to the
 parallel to can-
 the
 vas and can be tilted if needed. Normally, only the upper or lower rows would be tilted.
 canvas and can be tilted if needed. Normally, only the upper or lower rows would be tilted.

 The resolution
 The resolutionofofthe
 thefinal
 final gigapixel
 gigapixel image,
 image, achieved
 achieved withwith
 thosethose methods,
 methods, will de-
 will depend
 on several factors, such as the camera sensor resolution, the lens focal length, and and
 pend on several factors, such as the camera sensor resolution, the lens focal length, the
 the distance
 distance between
 between the canvas
 the canvas and and the camera,
 the camera, as will
 as will be discussed
 be discussed later.later.

 2.2.
 2.2. Illumination
 Illumination Set
 Set
 When
 When capturing
 capturing paintings
 paintings there
 there are
 are several
 several approaches
 approaches to to illuminate
 illuminate the the artwork
 artwork
 properly.
 properly. The
 The simplest
 simplest wayway isis to
 to use
 use the
 themuseum’s
 museum’s own own illumination
 illumination whenwhen the the capture
 capture
 process
 process is
 is carried
 carried outout in
 in situ.
 situ. Nevertheless,
 Nevertheless, for for aa uniform
 uniform and and colour-accurate
 colour-accurate resultresult itit
 would
 would bebe advisable
 advisable to to use
 use dedicated
 dedicated andand controlled
 controlled light
 light sources.
 sources. InIn this
 this research,
 research, there
 there
 were
 were used
 used two
 two Mettle
 Mettle 2828 inch
 inch soft
 soft boxes
 boxes lights,
 lights, providing
 providing aa colour
 colour temperature
 temperature of of 5500
 5500 KK
 and a CRI (Colour Rendering Index) greater
 and a CRI (Colour Rendering Index) greater than 90. than 90.
 The
 The soft
 soft boxes
 boxes must
 must be
 be placed
 placed strategically
 strategically toto avoid
 avoid specular
 specular reflections
 reflectionson onthe
 thecanvas
 canvas
 and
 and to guarantee lighting uniformity, so the lighting scheme proposed in Figure 9 is9ad-
 to guarantee lighting uniformity, so the lighting scheme proposed in Figure is
 advisable.
 visable. TheThe lights
 lights mustmust be placed
 be placed outsideoutside the family
 the family of anglesof zone,
 anglesaszone, as otherwise
 otherwise specular
 specular reflections
 reflections may occur. may occur.

J. Imaging 2021, 7, x FOR PEER REVIEW 8 of 20
J. Imaging 2021, 7, 156 8 of 20
 J. Imaging 2021, 7, x FOR PEER REVIEW 8 of 20

 Figure 9. Lighting scheme to illuminate the artwork uniformly. In order to avoid specular reflec-
 tions on
 Figure
 Figure 9.the
 9. canvas,
 Lighting
 Lighting the light
 scheme
 scheme to sources must
 to illuminate
 illuminate thebe
 the placeduniformly.
 artwork
 artwork outside ofIn
 uniformly. the
 In volume
 order
 order defined
 totoavoid
 avoid by thereflections
 specular
 specular family
 reflec-
 of
 on angles.
 tions
 theon the canvas,
 canvas, thesources
 the light light sources must
 must be be placed
 placed outsideoutside of the volume
 of the volume defineddefined by theof
 by the family family
 angles.
 of angles.
 When using
 When using thethe multi-viewpoint
 multi-viewpoint capture technique, it is mandatory
 mandatory to move move thethe
 lighting
 lighting
 When system
 systemusing along
 the with
 along with the
 the camera,
 camera, so
 multi-viewpoint a special
 capture mount was
 technique, it isdesigned
 mandatory to attach
 to movethe soft
 the
 boxes
 boxes to
 lighting the
 the tripod
 tosystem tripod (Figure
 with 10).
 (Figure
 along Additionally,
 10).camera,
 the a sliding
 so a special system
 mount was for the tripod’s
 designed which
 feet,the
 to attach which
 soft
 consisted
 consisted
 boxes to thein
 inatripod
 atwo-wheeled
 (Figure aluminium
 two-wheeled aluminium
 10). bar, was
 bar,
 Additionally, a also
 was constructed
 also
 sliding constructed
 system for to the
 facilitate thefeet,
 to tripod’s
 facilitate movement
 the which
 move-
 of the of
 ment set.the
 consisted in set.
 a two-wheeled aluminium bar, was also constructed to facilitate the move-
 ment of the set.

 Figure 10.
 Figure 10. Lighting
 Lighting mount
 mount to
 to attach
 attach the
 the lighting
 lighting system
 system to
 to the
 the tripod.
 tripod.
 Figure 10. Lighting mount to attach the lighting system to the tripod.
 2.3.
 2.3. Camera
 Camera Settings
 Settings and
 and Colour
 Colour Accuracy
 Accuracy
 2.3. Camera Settings
 ItItisisadvisable and
 advisabletotouseColour
 usean Accuracy
 anintermediate
 intermediatediaphragm
 diaphragm aperture
 aperturevalue
 valueto to
 avoid thethe
 avoid diffraction
 diffrac-
 phenomenon
 tion phenomenon whilewhile
 It is advisable tomaintaining
 usemaintaininggoodgood
 an intermediate depth of field.
 depth
 diaphragm Setting
 of field. the
 Setting
 aperture camera
 the
 value to f8
 camera
 to avoid to would be
 f8diffrac-
 the would
 advisable
 tion phenomenonas it is usually
 while the golden number
 maintaining for maximizing
 good depth imagethe
 of field. Setting sharpness. The
 camera to f8 camera
 would
 speed will depend on the lighting conditions. Slow speeds will not be a problem since a
 tripod is used.

J. Imaging 2021, 7, x FOR PEER REVIEW 9 of 20

J. Imaging 2021, 7, 156 9 of 20
 be advisable as it is usually the golden number for maximizing image sharpness. The cam-
 era speed will depend on the lighting conditions. Slow speeds will not be a problem since
 a tripod is used.
 The
 The camera
 camera ISO ISO value
 value must
 must be at at the
 the lower
 lower setting
 setting to avoid noisy pictures. An ISO
 100
 100value
 value isis advisable.
 advisable. RAWRAW Image
 Image file format
 format should be used instead of JPG in other other to
 to
 maximize the
 maximize the dynamicdynamic range and to properly adjust the white balance and
 balance and to apply
 apply anan
 accurate
 accurate colour
 colour profile
 profile during
 during thethe digital
 digital developing
 developing process.
 When
 When using
 using telephoto
 telephoto lenses,
 lenses, it is advisable to shoot the the camera
 camera remotely,
 remotely, even whenwhen
 aa tripod
 tripod is
 is used.
 used. For this reason,
 For this reason, itit is
 is recommended
 recommended using using aaremote
 remoteshooter
 shooteror,or,even
 evenbetter,
 better,
 aa mobile
 mobile device
 device or or notepad
 notepad with
 with aa compatible
 compatible application
 application for for shooting
 shooting the
 the camera.
 camera. In In this
 this
 research,
 research,aa notebook
 notebook with with the
 the free
 free software
 software EOSEOS Utility
 Utility byby Canon,
 Canon, waswas used
 used toto control
 control the
 the
 camera via Wi-Fi.
 camera via Wi-Fi.
 In
 In order
 order to to capture
 capture thethe colours
 colours of of the
 the artwork accurately, aa Xrite
 artwork accurately, Xrite ColorChecker
 ColorChecker chart chart
 was used to create a colour profile from a picture of the chart, which
 was used to create a colour profile from a picture of the chart, which was illuminated by was illuminated
 by
 thethe scene
 scene lights
 lights (Figure
 (Figure 11). colour
 11). The The colour
 profileprofile was generated
 was generated by the ColorChecker
 by the ColorChecker Camera
 Camera
 CalibrationCalibration
 softwaresoftware by Xrite
 by Xrite and it canand it canwith
 be used be used
 RAW with RAW developers,
 developers, such as thesuch as
 open-
 the opensource
 source RawTherapee RawTherapee
 [20], as well[20], as well
 as with otheras commercial
 with other commercial
 software, such software,
 as Adobe such
 Cam-as
 Adobe
 era Raw Camera
 or Adobe RawLightroom
 or Adobe [21].
 Lightroom [21]. The
 The picture of thepicture
 chart ofcanthe chart
 also can also
 be used as abe used as
 reference
 atoreference
 adjust the proper white balance of all the shots at once, by using the middle greymiddle
 to adjust the proper white balance of all the shots at once, by using the patch
 grey
 of thepatch
 chartoftothe chart to any
 neutralize neutralize
 colour any colour predominance.
 predominance.

 Figure11.
 Figure 11.Reference
 Referencepicture
 picturewith
 withthethe X-Rite
 X-Rite ColorChecker
 ColorChecker chartchart
 thatthat
 will will
 help help
 to settothe
 setproper
 the proper
 white
 white balance and to create a specific colour profile for the lighting conditions.
 balance and to create a specific colour profile for the lighting conditions.

 2.4.
 2.4. Generating
 Generating the
 the Gigapixel
 Gigapixel Image
 In
 In order
 ordertotojoin
 jointhe setset
 the of pictures obtained
 of pictures obtainedwithwith
 any any
 of theofaforementioned techniques,
 the aforementioned tech-
 aniques,
 stitching softwaresoftware
 a stitching such as Hugin
 such asorHugin
 PTGUIorcan be used.
 PTGUI As used.
 can be mentioned before, Hugin
 As mentioned before,is
 aHugin
 powerful is a opensource stitching software
 powerful opensource stitchingthat can even
 software thatcalculate
 can eventhe camera the
 calculate translation
 camera
 between
 translation shots, when the
 between multi-viewpoint
 shots, technique is used.
 when the multi-viewpoint The only
 technique drawback
 is used. of Hugin
 The only draw- is
 that is slow when compared to the commercial alternative PTGUI, which
 back of Hugin is that is slow when compared to the commercial alternative PTGUI, which is incredibly fast.
 The workflow
 is incredibly fast. with the stitching software is very simple. Initially, the set of pictures
 are analysed to automatically
 The workflow detect homologous
 with the stitching points
 software is very in theInitially,
 simple. overlapping
 the setzones of the
 of pictures
 shots. Then, the
 are analysed optimizer algorithm
 to automatically detect uses all thesepoints
 homologous pointsinto the
 calculate the spatial
 overlapping zonesposition
 of the
 of every picture, which is defined by the pitch, yaw, and roll angles. In addition to this, the
 camera translations, and the parameters to correct the radial distortion, caused by the lens,
 are also computed. The final step allows rendering the set of pictures into a whole image at

J. Imaging 2021, 7, 156 10 of 20

 gigapixel resolution. The maximum final resolution will depend on the number of shots
 and their original resolution. That is estimated by the software so as to avoid incremental
 interpolations.
 When using the multi-viewpoint technique, it is possible to assemble the gigapixel
 image with an automated SfM photogrammetry software, such as Agisoft Metashape. As
 mentioned before, using this kind of software would require a greater overlapping between
 adjacent shots and it is much more time consuming. This option would be interesting only
 when documenting volumetric artworks, such as altar pieces. The workflow for this kind
 of software begins with the determination of the camera positions from the analysis of the
 homologous points and the sparse point cloud. Then, the dense cloud of the model can
 be computed, and can be later triangulated to form a polygonal mesh. This mesh can be
 texturized by projecting the original pictures onto the mesh. Finally, an orthophoto of the
 3D textured model can be generated at gigapixel resolution.
 It should be mentioned that SfM photogrammetric techniques would not provide an
 absolutely accurate three-dimensional restitution of those subtle surface details, but since
 our main objective is obtaining an orthophoto of the painting, it would provide excellent
 results. In order to render fine surface properties of the painting, reflectance transformation
 imaging (RTI) would be very useful, as well as other much more expensive 3D scanning
 techniques [22].

 2.5. Canvas Measurement
 Knowing the real dimension of the paintings is important to provide an accurate scale
 of the gigapixel image. Normally the dimensions of the artwork are specified on the wall
 labels, but they are approximated. A Topcon Image IS robotic total station was used for
 taking the measurement of the canvases without touching the artwork. This total station
 has a reflector-less measurement capability between 1.5 and 250 m, providing a short-range
 precision of ±5 mm (MSE), and an angular precision of 0.3 mgon. The instrument was
 placed in front of the painting, and, with the help of the telescopic sight, the corners of the
 canvas were identified. The coordinates of these points were obtained using TopSurv, a
 software that comes with the total station.

 3. Results
 In this section, some examples, which were captured using the described techniques,
 will be exposed. All the paintings belong to the Museo de Bellas Artes of Valencia (Spain).

 3.1. Single-Viewpoint Capture
 The first example is focused on the artwork entitled La Santa Cena, painted by Juan
 de Juanes in 1534. This artwork measures 92.3 cm tall × 85.8 cm wide. Due to its reduced
 size, it was shot using the Single-Viewpoint Technique. The camera was placed in a central
 position with respect to the canvas and at a distance of 2 m. In this manner, the perimeter
 parts were not too oblique to the camera sensor, thus preserving the image sharpness
 (Figure 12). All the zones had to be inside depth of field limits provided by the telephoto
 lens. The diaphragm was set to f8, the ISO value to 100, and the shutter speed was set to
 2 s in accordance with the lighting conditions.

tion software, which is provided by the manufacturer of the colour chart.
 After the raw development process, the mosaic was joined using Hugin, a GPL-li-
 censed stitching software, which generated the final image (Figure 13b), with a resolution
 of 26,511 × 28,520 px, that is, 756 megapixels (0.76 gigapixels).
 It is interesting to express the resolution relative to a measurement unit, so this value
J. Imaging 2021, 7, 156 11 of 20
 would be independent of the artwork size. This parameter is known as pixel density and
 will allow an objective comparison of the level of detail, acquired in each case. The pixel
 density, expressed in pixel per inch, was 787 ppi in this case.

 Figure 12. Single-viewpoint capture technique. Viewpoint position in the capture of the Santa Cena
 Figure 12. Single-viewpoint capture technique. Viewpoint position in the capture of the Santa Cena
 by Juan de Juanes (1534). Museo de Bellas Artes de Valencia (Spain).
 by Juan de Juanes (1534). Museo de Bellas Artes de Valencia (Spain).

 A total of 90 photographs were taken by rotating the camera around its centre of
 perspective, in steps of 2.5 degrees (horizontal rotation), and 4 degrees (vertical rotation).
 That resulted in a mosaic of 10 columns and 9 rows, with an overlapping between adjacent
 J. Imaging 2021, 7, x FOR PEER REVIEW 12 of 20
 pictures greater than 30% (Figure 13a).

 Figure13.
 Figure 13.(a)
 (a) Set
 Set of
 of 90
 90 pictures
 pictures taken
 taken during
 during the
 the gigapixel
 gigapixel capture.
 capture. (b)
 (b) Stitched
 Stitched gigapixel
 gigapixel Image.
 Image.

 AFigure 14 shows
 reference a fragment
 picture, with theofX-Rite
 the resulting
 Colourimage
 Checkerin which
 Chart,the level
 was of detail
 taken can be
 to generate
 anoticed.
 specificThe pixel
 colour density
 profile foristhe
 not huge because
 lighting of the
 conditions of long distance
 the scene, andbetween
 to adjustthe
 thecamera
 white
 and the in
 balance canvas. This distance
 the RAW developmentwas increased to preventIn
 process precisely. thethis
 losscase,
 of sharpness
 only the in the decen-
 lights of the
 tred shots
 museum dueused
 were to the
 asobliquity between
 the lighting the canvas and the camera sensor. Nevertheless,
 source.
 a greater resolution could have been achieved if the proposed parallel multi-viewpoint
 capture method had been used, so the camera would have been placed almost at the min-
 imum focusing distance of the lens (1.5 m instead of 2 m).

J. Imaging 2021, 7, 156 12 of 20

 The development process was carried out using the opensource software RAW Ther-
 apy. There were generated 90 files in TIF format with 32 bits per channel, which were
 perfectly balanced with the colour profile created by the ColorChecker Camera Calibration
 software, which is provided by the manufacturer of the colour chart.
 After the raw development process, the mosaic was joined using Hugin, a GPL-
 licensed stitching software, which generated the final image (Figure 13b), with a resolution
 of 26,511 × 28,520 px, that is, 756 megapixels (0.76 gigapixels).
 It is interesting to express the resolution relative to a measurement unit, so this value
 would be independent of the artwork size. This parameter is known as pixel density and
 will allow
 Figure anSet
 13. (a) objective comparison
 of 90 pictures of thethe
 taken during level of detail,
 gigapixel acquired
 capture. in each
 (b) Stitched case. The
 gigapixel pixel
 Image.
 density, expressed in pixel per inch, was 787 ppi in this case.
 Figure
 Figure 1414 shows
 shows aa fragment
 fragment of of the
 the resulting
 resulting image
 image inin which
 which the level of
 the level of detail
 detail can
 can be
 be
 noticed.
 noticed. The
 Thepixel
 pixeldensity
 densityisisnot
 nothuge
 huge because
 becauseof of
 thethe
 long distance
 long distancebetween the camera
 between the cameraand
 the
 andcanvas. This This
 the canvas. distance was was
 distance increased to prevent
 increased the loss
 to prevent of sharpness
 the loss of sharpnessin the decentred
 in the decen-
 shots due to
 tred shots the
 due toobliquity between
 the obliquity the the
 between canvas
 canvasandand
 thethe
 camera
 camera sensor.
 sensor.Nevertheless,
 Nevertheless,a
 greater resolution could have been achieved if the proposed parallel
 a greater resolution could have been achieved if the proposed parallel multi-viewpoint multi-viewpoint
 capture
 capture method
 method had hadbeen
 beenused,
 used,soso thethe camera
 camera would
 would havehave
 beenbeen
 placedplaced
 almostalmost
 at theatmin-
 the
 minimum focusing distance of the lens (1.5 m instead
 imum focusing distance of the lens (1.5 m instead of 2 m). of 2 m).

 Figure 14.
 Figure 14. Fragment
 Fragment of
 of the
 the resulting
 resulting gigapixel
 gigapixel image.
 image.

 3.2.
 3.2. Parallel
 Parallel Multi-Viewpoint
 Multi-Viewpoint Capture
 The parallel multi-viewpoint capture method will produce better image quality than
 the previous one since the camera sensor is always parallel to the canvas. Consequently,
 all the zones of the pictures would be perfectly focused. Nevertheless, this approach is
 more complicated, so the camera must be moved along with the lights every shot, but it is
 worth it.
 This technique was put into practice in the next example to capture the painting
 entitled Virgen de la Leche, painted by Bartolomé Bermejo around 1478. The canvas size is
 quite small, measuring 49.1 cm tall × 34.5 cm wide.
 In this case, the previously described lighting system was used, which consisted of
 two soft boxed attached to the tripod (Figure 10). The lamps of the museum were turned-off
 so as not to mix different light sources.

more complicated, so the camera must be moved along with the lights every shot, but it
 is worth it.
 This technique was put into practice in the next example to capture the painting en-
 titled Virgen de la Leche, painted by Bartolomé Bermejo around 1478. The canvas size is
J. Imaging 2021, 7, 156 quite small, measuring 49.1 cm tall × 34.5 cm wide. 13 of 20
 In this case, the previously described lighting system was used, which consisted of
 two soft boxed attached to the tripod (Figure 10). The lamps of the museum were turned-
 off so as not to mix different light sources.
 The
 The photographic
 photographic set set must
 must be
 be stabilized
 stabilized after
 after changing the camera
 changing the camera position,
 position, so
 so the
 the
 flexibility of the light stand, which is attached to the tripod, can lead to slightly
 flexibility of the light stand, which is attached to the tripod, can lead to slightly blurred blurred
 shots
 shots due
 due toto small
 small vibrations
 vibrations transmitted
 transmitted toto the
 the camera. This is
 camera. This is controlled
 controlled by by monitoring
 monitoring
 the camera with the laptop. The live view image, which is received
 the camera with the laptop. The live view image, which is received via Wi-Fi from via Wi-Fi from the
 the
 camera, is magnified 10 × to detect any slight vibration and to focus the
 camera, is magnified 10× to detect any slight vibration and to focus the canvas very accu- canvas very
 accurately.
 rately. Once Once the image
 the image is completely
 is completely steady
 steady and focused,
 and focused, the camera
 the camera is remotely
 is remotely shot
 shot (Fig-
 (Figure
 ure 15). 15).

 Figure 15.
 Figure 15. Remote
 Remote control
 control of
 of the
 the camera
 camera with
 with the
 the laptop
 laptop via
 via Wi-Fi.
 Wi-Fi.

 We took
 We took 20
 20 shots
 shots to
 to cover
 cover the
 the entire
 entire canvas
 canvas obtaining a mosaic of 5 columns
 columns and
 and 44
 rows with
 rows with an
 an overlapping
 overlapping greater
 greater than
 than 30%
 30% (Figure 16a). The camera was placed at 1.5 m
 from the
 from the canvas
 canvas and
 and moved
 moved horizontally
 horizontally to cover the upper row. Then, the camera was
 moved downwards vertically
 moved verticallytotoinitiate
 initiatethe
 thesecond
 secondrow
 rowand so on.
 and A measuring
 so on. tape tape
 A measuring was
 attached to the floor, parallel to the canvas, to serve as a reference for the translation
 was attached to the floor, parallel to the canvas, to serve as a reference for the translation move-
 ment. The diaphragm
 movement. was set
 The diaphragm wasto set
 f8, the ISO
 to f8, thevalue to 100,toand
 ISO value 100,the
 andshutter speed speed
 the shutter was set to
 was
 1/20
 set toof second
 1/20 according
 of second to the to
 according lighting conditions.
 the lighting conditions.

J. Imaging 2021, 7, 156 14 of 20
 J. Imaging 2021, 7, x FOR PEER REVIEW 14 of 20
 J. Imaging 2021, 7, x FOR PEER REVIEW 14 of 20

 Figure 16.
 16. (a) Set
 Set of 20
 20 pictures taken
 taken during the gigapixel capture. (b) Stitched Gigapixel Image.
 Figure 16. (a)
 Figure (a) Set of
 of 20 pictures
 pictures taken during the gigapixel capture. (b) Stitched Gigapixel Image.
 Image.

 The
 The raw
 The raw developing
 rawdeveloping process
 developingprocess
 processwas was carried
 wascarried
 carriedoutout
 outinin the
 inthe same
 thesame
 sameway way
 wayasas
 as the
 the
 the previous
 previous
 previous exam-
 example.
 exam-
 ple.
 ple. Nevertheless, the shots were stitched using PTGUI instead of Hugin as it is much
 Nevertheless,
 Nevertheless, the the
 shots shots
 were were
 stitched stitched
 using using
 PTGUI PTGUI
 instead instead
 of Hugin of
 as Hugin
 it is as
 much it is
 faster. The
 much
 faster.
 resultingThe resulting image
 (Figure 16b) (Figure
 has a 16b)
 final has a final
 resolution resolution
 of 14,183 × of 14,183 ×
 faster. The resulting image (Figure 16b) has a final resolution of 14,183 × 20,199 px, whicha
 image 20,199 px, 20,199
 which px, which
 provides
 provides
 pixel
 provides a
 a pixel
 density pixel density
 of 1044 ppi.of
 density 1044
 This
 of 1044 is ppi.
 much
 ppi. This is
 is much
 better
 This than in
 much better than
 than in
 the previous
 better in the previous
 case,
 the case,
 since the
 previous case, since
 camera
 since
 the
 the camera could be placed closer to the canvas. As mentioned, there is no problem with
 could camera
 be could
 placed be placed
 closer to thecloser
 canvas.to the
 As canvas. As
 mentioned, mentioned,
 there is nothere is
 problem no problem
 with depth
 withof
 depth
 field of
 when field when
 using this using
 method.this method.
 Figure 17 Figure
 shows a 17 shows
 cropped a cropped
 fragment
 depth of field when using this method. Figure 17 shows a cropped fragment of the gi- of fragment
 the of
 gigapixel the gi-
 image
 gapixel
 centred image
 gapixel on
 image centred
 the eye on
 of the
 centred the
 the eye
 onVirgin.eye ofof the
 the Virgin.
 Virgin.

 Figure 17. Fragment of the resulting gigapixel image.
 Figure 17.
 Figure 17. Fragment
 Fragment of
 of the
 the resulting
 resulting gigapixel
 gigapixel image.
 image.

 3.3.
 3.3. Tilted
 Tilted Multi-Viewpoint
 Multi-Viewpoint Capture
 Capture

The last example shows the gigapixel capture of the artwork entitled Martirio de San
 Bartolomé, painted by Luca Giordano around 1650. The canvas size is 136 cm tall × 95 cm
J. Imaging 2021, 7, 156 wide, which is a bit large to use the single-viewpoint method properly, so image sharp- 15 of 20
 ness would be poor in some zones of the boundary shots. Moreover, the upper and lower
 borders of the painting exceeded the maximum and minimum height of the tripod, so the
 tilted multi-viewpoint capture method was chosen in this case.
 3.3. Tilted Multi-Viewpoint Capture
 We took 90 shots, forming a mosaic of 10 columns and 9 rows (Figure 18a). Most of
 The lastwhere
 the pictures example shows
 parallel to the gigapixel
 the canvas. capture
 Only of the
 the shots artwork entitled
 corresponding to theMartirio
 upper and de
 San Bartolomé, painted by Luca Giordano around 1650. The canvas size
 lower rows were slightly tilted vertically to cover the entire artwork, which is negligible is 136 cm tall ×
 95 cm wide, which is a bit large to use the single-viewpoint method
 for the image quality. The distance from the camera to the canvas was set to 1.9 m. The properly, so image
 sharpness wouldwere:
 camera settings be poor in somef8,
 diaphragm zones
 ISO of
 100,the boundary
 and shots. 1/5
 camera speed Moreover,
 of second,theaccordingly
 upper and
 lower borders of the
 to the lighting conditions.painting exceeded the maximum and minimum height of the tripod,
 so theAs tilted multi-viewpoint capture method was chosen in this case.
 in the previous cases, the RAW developing process was carried out with RAW-
 We took
 Therapee 90 shots,
 to apply the forming
 proper whitea mosaic of 10and
 balance columns and 9colour
 the specific rows (Figure 18a).
 profile to all Most
 the pic-of
 the pictures where parallel to the canvas. Only the shots corresponding
 tures at once. The stitching was carried out with PTGUI, which provided a gigapixel im- to the upper and
 lower rows were
 age (Figure slightly
 18b) with tilted
 a total vertically
 resolution of to cover× the
 30,884 entire
 44,228 artwork,
 px (1.36 which isInnegligible
 gigapixels). this case,
 for
 the pixel density was 826 ppi, something lower than in the previous caseset
 the image quality. The distance from the camera to the canvas was duetoto1.9
 them. The
 larger
 camera settings were: diaphragm f8, ISO 100, and camera speed 1/5 of
 distance between the camera and the canvas. Figure 19 shows a cropped fragment of the second, accordingly
 to the lighting
 gigapixel imageconditions.
 focused on the saint’s mouth.

 Figure18.
 Figure 18.(a)
 (a) Set
 Set of
 of 90
 90 pictures
 pictures taken
 taken during
 during the
 the gigapixel
 gigapixel capture.
 capture. (b)
 (b) Stitched
 Stitched gigapixel
 gigapixel image.
 image.

 As in the previous cases, the RAW developing process was carried out with RAWTher-
 apee to apply the proper white balance and the specific colour profile to all the pictures
 at once. The stitching was carried out with PTGUI, which provided a gigapixel image
 (Figure 18b) with a total resolution of 30,884 × 44,228 px (1.36 gigapixels). In this case, the
 pixel density was 826 ppi, something lower than in the previous case due to the larger
 distance between the camera and the canvas. Figure 19 shows a cropped fragment of the
 gigapixel image focused on the saint’s mouth.

J. Imaging 2021, 7, 156 16 of 20
 J. Imaging 2021, 7, x FOR PEER REVIEW 16 of 20

 Figure 19.
 Figure 19. Fragment
 Fragment of
 of the
 the resulting
 resulting gigapixel
 gigapixel image.
 image.

 4.
 4. Discussion
 Discussion
 The
 The new
 new proposed
 proposed multi-viewpoint
 multi-viewpoint gigapixel capture techniques, compared to
 techniques, compared to the
 conventional single-viewpoint capture technique, improve the
 conventional single-viewpoint capture technique, improve the image quality of the result- image quality of the re-
 sulting gigapixel image. The parallel multi-viewpoint gigapixel
 ing gigapixel image. The parallel multi-viewpoint gigapixel capture technique is, in gen- capture technique is, in
 general,
 eral, most most recommended,
 recommended, so so there
 there is is
 nono lossofofsharpness
 loss sharpnessproduced
 producedby bynarrow
 narrow depth
 depth ofof
 field
 field provided
 provided by by telephoto
 telephoto lenses.
 lenses. The novel multi-viewpoint capture techniques techniques implyimply
 using
 using different
 different shooting
 shooting methods
 methods and moving the light system along with the camera camera to to
 avoid changing specular reflections during the
 avoid changing specular reflections during the capture process. capture process.
 Nevertheless,
 Nevertheless, the the single-viewpoint
 single-viewpoint capturecapture technique
 technique can can be
 be also
 also aa proper
 proper method
 method
 when the depth of field provides an acceptable sharpness
 when the depth of field provides an acceptable sharpness in the whole picture. in the whole picture.
 TheThe
 ac-
 acceptable sharpness would depend on the focal lens, the diaphragm
 ceptable sharpness would depend on the focal lens, the diaphragm aperture value, the aperture value, the
 focus
 focus distance,
 distance, and andthe thecamera
 camerasensor
 sensor size.
 size. It can
 It can be calculated
 be calculated by using
 by using a free aDOFfreecalcu-
 DOF
 calculator
 lator such such as DOFMASTER
 as DOFMASTER [23]. [23].
 From
 Fromthethe experience
 experience acquired
 acquired in in this research, there
 this research, there are
 are some
 some practical
 practical considerations
 considerations
 that can be very useful for future researchers, which will
 that can be very useful for future researchers, which will be discussed here. be discussed here.
 When
 When documenting
 documenting artworksartworks withwith gigapixel
 gigapixel resolution,
 resolution, the the pixel
 pixel density
 density isis an
 an im-
 im-
 portant aspect to bear in mind when planning the photographic
 portant aspect to bear in mind when planning the photographic session. A pixel density session. A pixel density
 range
 range between
 between 600 600 ppi
 ppi and
 and about
 about 1000
 1000 ppi
 ppi was
 was established
 established to to guarantee
 guarantee an an excellent
 excellent level
 level
 of
 of detail
 detail ofof the
 the paintings.
 paintings. The The maximum
 maximum reachable
 reachable pixelpixel density value is
 density value is always
 always imposed
 imposed
 by the photographic equipment, so it depends on the lens
 by the photographic equipment, so it depends on the lens focal length, the camerafocal length, the camera distance,
 dis-
 and the camera sensor resolution and
 tance, and the camera sensor resolution and size. size.
 In
 In order
 order to to maximize
 maximize the the pixel
 pixel density,
 density, itit would
 would be be advisable
 advisable to to use long focal
 use long focal lenses
 lenses
 and to place the camera as close to the painting as possible (the
 and to place the camera as close to the painting as possible (the minimum focusing dis-minimum focusing distance
 of the lens
 tance of thecan becan
 lens the bemostthelimiting factor).factor).
 most limiting
 The pixel density of the final gigapixel
 The pixel density of the final gigapixel image
 image cancanbebecalculated
 calculated easily
 easilyby by
 Equation
 Equation(1),
 which can be deduced geometrically:
 (1), which can be deduced geometrically:
 S r ∙· 
 FL∙ · U
 f
 =
 PD (1)
 D∙ · S
 w

 where,
 where,
 =PD = Pixel
 Pixel Density
 Density of theofImage
 the Image
 (ppc (ppc or ppi)
 or ppi)