1 Close-range photogrammetry enables documentation of environment-induced

2 deformation for architectural heritage

3

4 Jūratė Sužiedelytė-Visockienė1, Renata Bagdžiūnaitė1, Naglis Malys2, Vida Malienė3,4*

5

1
6 Department of Geodesy and Cadastre, Vilnius Gediminas Technical University, Vilnius 10223,
7 Lithuania
2
8 School of Life Sciences, University of Warwick, Coventry CV4 7AL, United Kingdom
3
9 School of the Built Environment, Liverpool John Moores University, Liverpool L3 3AF, United
10 Kingdom
4
11 Institute of Land Management and Geomatics, Faculty of Water and Land
12 Management, Aleksandras Stulginskis University, Universiteto 10, Akademija, Kaunas 53361,
13 Lithuania
14
*
15 Corresponding author; e-mail: v.maliene@ljmu.ac.uk; phone no: +44 (0)1512312854
16
 1 Abstract
2 Deformation, damage and permanent loss of heritage assets due to various physical and
3 environmental factors has always been a major problem. As the availability of funds for
4 conservation and restoration is limited, the digital documentation of heritage objects and monitoring
5 of environment-induced deformations are increasingly important for cultural heritage preservation.
6 Our study elucidates developments in the digital image capturing and processing for recording
7 architectural heritage objects focusing on the digital camera calibration, close-range imaging, and
8 photogrammetric modelling of complex structures using image matching techniques. A particular
9 consideration in this paper is given to the ortho-photographic image compiling and accuracy
10 assessment procedure. The practicality of the methodology is demonstrated by applying
11 photogrammetric system PhotoMod for documentation of decorative elements in Uzutrakis manor,
12 a national heritage site in Trakai, Lithuania.
13

14 Keywords: heritage, close-range photogrammetry, surface and geometric deformations, ortho-

15 photographic model, Uzutrakis manor
16

2
 1 1. Introduction
2 Cultural heritage is an invaluable asset of a human culture and creativity. A large proportion of it is
3 formed of architectural objects such as buildings and other physical structures that are historically
4 inherited from the past and sought to be sustained for the future. These objects distinguish
5 themselves by their historic, architectural, and technological significance. Regrettably, the
6 architectural heritage suffers from an irreversible deformation and damage caused by physical
7 factors of the natural environment such as aging, humidity, dust, weathering, pressure and others
8 (Castellini et al., 2008; López-Aparicio and Grašiene, 2013; Varas-Muriel et al., 2014), as well as
9 human activities and negligence (Norrström, 2013). A deterioration of the artwork can be monitored
10 by recording surface and geometric deformations (Hinsch et al., 2007; Castellini et al., 2008;
11 Remondino et al., 2011).
12 Often, original architectural drawings, which are necessary for heritage object restoration,
13 are not available or archaic and low quality. Work aimed at planning the restoration of architectural
14 heritage demands an up-to-date and accurate documentation on architectural and structural
15 characteristics, geometric shapes and materials. Therefore, the documentation and deformation
16 monitoring of complex heritage objects are of a vital importance (Chen and Romice, 2009;
17 Mazzanti, 2002; Nelle, 2009; Pickard, 2002; Shipley and Reeve, 2010).
18 Currently used strategies for architectural heritage documentation are based on spatial data
19 acquisition techniques such as tacheometry (conventional surveying), laser scanning and
20 photogrammetry (Grussenmeyer et al., 2008). Laser scanning and photogrammetric methods or
21 their combination can generate an accurate documentation of the cultural heritage with high-
22 resolution models of physical structures representing different perspectives of view (Al-kheder et
23 al., 2009; Beraldin et al., 2000; Campana and Remondino, 2007; Guarnieri et al., 2013; Han, 2012;
24 Kersten et al., 2009; Martorelli et al., 2014; Remondino, 2011; Remondino et al., 2009; Yastikli et
25 al., 2007; Yilmaz et al., 2008; Zheng et al., 2012). However, seeking to reduce costs or due to a lack
26 of technical skills, architectural conservators involved in heritage conservation still often use
27 traditional manual methods, despite that their application may result in a reduced accuracy or even
28 complete loss of information in the process of data transmission from the original object to the
29 theoretical model (Yilmaz et al., 2007).
30 The laser scanning technology and digital image photogrammetry provides possibility for
31 three-dimensional (3D) digitalization of architectural structures and their elements, which enables a
32 digital documentation and continuous monitoring of the spatial information at different time periods
33 (Al-kheder et al., 2009; Remondino, 2011; Pesci et al., 2013; Tapete et al., 2013). Both
34 technologies generate sets of data points that can be transformed into point clouds in 3D space,
35 which are required for the creation of two- or three-dimensional models (Akman et al., 2010; Gruen

3
 1 and Akca, 2005; Yastikli et al., 2007). Although acquiring data for point clouds using laser
2 scanning is a faster technique in the field of surveying, the arising issues such as affordability,
3 restricted accessibility of hidden-enclosed building areas, limited portability, required special
4 expertise and other factors reduce the possibility to use the laser scanning in favour of
5 photogrammetric methods (Arias et al., 2006; Chandler et al., 2005, 2007; Martínez et al., 2013;
6 Sužiedelyte-Visockiene et al., 2011).
7 As a result of the fast development of user-friendly close-range photogrammetric software
8 packages such as ImageMaster (Topcon Positioning Systems, Inc., Livermore, USA),
9 PhotoModeler (Eos Systems Inc., Vancouver, Canada), PhotoMod (Racurs, Moscow, Russia),
10 ShapeCapture (ShapeQuest Inc., Nepean, Canada) and others (Adrov et al., 1995; Aguilar et al.,
11 2005; Deng and Falg, 2001; Sužiedelyte-Visockiene et al., 2011), there is evident increase in the
12 use of 3D image-based digitalisation methods in the field of cultural heritage (Barazzetti et al.,
13 2011; Remondino et al., 2009; Yilmaz et al., 2008). With a variety of different approaches being
14 available, digital models can provide highly accurate both 2D and, when required, 3D drawings,
15 which contribute to the documentation of complex buildings, detailed decorative elements and other
16 architectural structures (De Reu et al., 2013; Tiano et al., 2008; Barbetti et al., 2013). In contrast to
17 the theoretical or simulated models, these drawings are able to reproduce accurately the morphology
18 of the object capturing differences of irregular elements and multiform geometry (Arias et al., 2007;
19 Remondino, 2011; Riveiro et al., 2011). A great deal of interest has been attracted to the 3D image-
20 based technology that combines the close-range photogrammetry with an automatic image matching
21 (AIM) and enables to create dense point clouds with 3D data from stereoscopic pair of images,
22 which can be captured using conventional digital cameras (Koutsoudis et al., 2013; Martínez et al.,
23 2013, Ortiz et al., 2013).
24 The paper reviews the pipeline for generation of ortho-photografic and 2D models using
25 digital close-range photogrammetry. It provides a detailed example of the process starting with the
26 calibration of the digital camera and leading to the generation of 2D models, which are required for
27 national heritage documentation in Lithuania. In this study, the close-range photogrammetry is used
28 to record and monitor the environment-induced deformation of the decorative elements on the
29 internal wall of Užutrakis palace (Deveikiene and Deveikis, 2011), which is a historical building
30 listed as a national heritage site. The palace has been built by order of Joseph Tyszkiewicz in 1898
31 – 1901 (LSE, 1983). The architect of palace was Warsaw architect Jozef Huss (1846-1904).
32 Buildings architectural features consist of few styles, historicism, neo-renaissance, and
33 neoclassicism. Interior was designed by following Ludwig XVI classicism.
34

35 2. Use of the close-range photogrammetry for documentation of the architectural heritage

4
 1 Image-based modelling using photogrammetry is regarded as one of the best techniques for
2 processing of image data, which provides accurate and detailed 3D information. This technique
3 allows determining the precision and reliability of data parameters from the measured image tie
4 points and control points (CP), which represent the location of matching/correspondence points in
5 two or several adjacent digital images and can be used to link images. Therefore, at least two images
6 are required, and then the 3D information can be derived by applying projective and perspective
7 geometry expressions.
8 The close-range photogrammetry is largely preferred over other techniques for data
9 acquisition and processing when working with architectural structures and elements, which have
10 small objects of irregular shape, for mapping applications and, in particular, for reconstruction of a
11 missing elements in the larger object or for documentation of surface and geometric deformations. It
12 is usually less cost demanding, with fewer time and location constrains and provides an enjoyable
13 experience of working in the team (Remondino and Rizzi, 2010). It can be employed in many
14 different applications relevant to the cultural heritage including documentation, conservation, digital
15 restoration, deformation analysis, monitoring, visualization and painting analyses (Arias et al., 2005;
16 2006; Fassi et al., 2011; Remondino et al., 2011; Yastikli, 2007; Yilmaz et al., 2007; 2008).
17 A substantial number of studies have reported on how the close-range photogrammetry can be
18 used to survey objects with different levels of complexity and arrangement suiting high quality
19 requirements and entailing only a few restrictions (Barazzetti et al., 2013; Barbetti et al., 2013;
20 Bitelli et al., 2011; Grussenmeyer et al., 2008; Ordóñez et al., 2010; Pérez-Gracia et al., 2011; Sanz
21 et al., 2010; Solla et al., 2012).
22 Currently available technology allows using terrestrial, aerial or satellite platforms (synthetic
23 aperture radar or lidar surface topography) for 3D information acquisition, some of which can then be
24 used in a typical photogrammetric pipeline starting with digital camera calibration and finishing with
25 ortho-photographic model generation (Fig. 1).
26

5
 1
2 Fig. 1. Pipeline for digital image capturing and photogrammetric processing. Digital image
3 capturing platform consists of six main elements. It includes camera calibration (element 1; error
4 values due to camera optics distortions are established), marking of control and tie points (element
5 2; coordinates of control points on the object are determined, which are used for geo-referencing,
6 element 3), positioning of targets (element 4; identification of targets for image acquisition) image
7 acquisition (element 5; overlapping images of the object are acquired), and image correction
8 (element 6; on the basis of camera calibration data the image is corrected). Photogrammetric
9 processing of the image can include eight different elements. These are following: orientation
10 (element 1; image orientation in respect to tie and control points are established), triangulation and
11 computing of X, Y, Z coordinates of points (element 2 and 3; X, Y, Z coordinates of points from
12 overlapping images are computed), creating of point cloud (element 4; on the basis of established
13 X, Y, Z coordinates the point cloud can be generated), and wire-frame model (element 5), which the
14 contributes to ortophoto generation (element 6) that forms the basis for high resolution 2D model
15 (element 7). Many elements of these elements in the photogrammetric processing must meet
16 accuracy and quality control requirements (element 8).
17

18 The photogrammetry involves steps that are performed in an automated way (e.g. geo-
19 referencing, image correction, camera calibration, orientation, ortophoto generation) with a little

6
 1 expert knowledge required or manually, where a high level of precision is critical (e.g. for image
2 processing of very small objects) (Sužiedelytė-Visockienė, 2013; Sužiedelytė-Visockienė et al.,
3 2014). Depending on the project type and quality requirements, a high level of professional expertise
4 can be required for the interaction with and intervention at specific stages of the modelling pipeline of
5 the photogrammetric process.
6 Manual procedures are still necessary for accurate element extraction from the satellite, aerial
7 or terrestrial image. Predominantly automated methods are becoming increasingly available to the
8 cultural heritage community. They can be used effectively for visualization, object-based navigation,
9 annotation or image browsing (Vergauwen and Van Gool 2006; Remondino, 2011). However, they
10 are often insufficient for the accurate heritage object reconstruction and documentation. Therefore, the
11 automation of image-based modelling of architectural structures and elements remains an important
12 research subject (Barazzetti et al. 2010; 2011; Oniga and Diac, 2013), as the expert intervention is
13 still necessary for the accurate camera calibration, accuracy and quality control of the
14 photogrammetric processing, geo-referencing and few other stages (see Fig. 1).
15

16 3. Image acquisition and photogrammetric processing

17 Professional-terrestrial digital cameras are commonly used for image acquisition by close-range
18 photogrammetry. They come in different forms and shapes containing CCD, CMOS or Live MOS
19 sensor, single or multiple and/or panoramic heads, frame or linear array, consumer, SLR-type or
20 industrial, low- or high-speed, etc (Remondino, 2011). Minimal requirement for terrestrial cameras
21 is to have a sensor with a resolution of at least 12 megapixels while high-end digital cameras span
22 to 60 megapixel sensors. Although panoramic linear array cameras are available on the market and
23 they can deliver a very high quality and resolution images with exceptional metrics, due of their
24 high cost, panoramic images frequently are generated by combining together set of overlapping
25 images acquired from a single point of view with a consumer range digital camera. Such a low-cost
26 solution is used not only for Google Street View, but is common close-range photogrammetry
27 applications focussed on documentation of the architectural heritage.
28 During the photogrammetric measurements, digital two-dimensional (2D) or three -
29 dimensional (3D) images of the object or its details, as well as ortho-photographic model, are
30 derived. This allows recording the status of the object with its visible surface and geometric
31 deformations. In most cases, such deformations could be recorded by manually performed
32 measurements. Therefore, it is appropriate to present the analysis of the digital photogrammetric
33 method instead. The object is usually imaged using calibrated professional camera with sensor of
34 high resolution. Obtained overlapping images are processed using specialised photogrammetric

7
 1 software. The following stages of photogrammetric measurements are essential in order to compile
2 the ortho-photographic model of images (see Fig. 1):
3  Interior orientation;
4  Relative orientations;
5  Calculation of triangulation;
6  Adjustment of accuracy for the ortho-photographic model creation.
7 While executing all the above mentioned processes it is necessary to monitor the accuracy of
8 derived results. The precision and reliability of photogrammetric process heavily depend on that.
9

10 4. Methodology
11 In this study the following initial preparations were performed as recommended by Kutut (2011):
12  Inspection and survey of the object and photo-fixation;
13  Photogrammetric measurements;
14  Analysis of the available documentation and historical research study records;
15  Geodetic measurements;
16  Analysis of composition of the object.
17 Objects of this survey were an internal wall of the chamber and associated wall ornaments in
18 Uzutrakis palace. The imaging of the wall and all related work was conducted in 2012. The
19 following preparatory tasks were performed at the site of this heritage object:
20  Marking and measurement of the geodetic control points with the geodetic equipment;
21  Imaging of the wall;
22  Image correction, due to the errors of the camera optics.
23 Photogrammetric processing of the images was performed as following:
24  Measurement of image central points (inner orientation);
25  Measurement of geodetic control points and tie points of the model by means of the
26 photogrammetric program (relative orientation);
27  Calculate of triangulation adjustment;
28  Stereo digitalization of the wall (determination of structural lines);
29  Compiling of the detailed ortho-photographic models;
30  Drawing of the details on the ortho-photographic models.
31 The measurements of the geodetic CP of the wall and ornaments as well as tie points
32 together with the triangulation adjustments were carried out by using the photogrammetric system
33 PhotoMod (Adrov et al., 1995). The processing of images in PhotoMod included the interior
34 orientation and relative orientation, the input and measurement of geodetic control points and tie

8
 1 points located in the overlapping areas between the adjusted images (RACURS, 2012).The accuracy
2 of the tie and CP point measurements was assessed on the basis of following criteria:
3 1. Correlation coefficient. On the basis of image quality, the user can determine the
4 acceptable value of the correlation coefficient. For contrast, clear and high quality images, the
5 correlation coefficient threshold was adjusted to 0.9–0.95, while for lower quality images this
6 threshold was reduced to 0.8;
7 2. Vertical parallax residual. The mean value of the vertical parallax was not allowed to
8 be greater than 50% of the matrix pixel size of the digital camera. Since the matrix pixel size of the
9 camera Canon EOS 1D Mark III was 6 µm, the mean value of the vertical parallax was not allowed
10 to be greater than 3 µm. The maximum error ( E max ) and root mean squared error (RMS) were
11 calculated using following equations:

12 E max  2  E mean , (1)

13 RMS = 2 ´ Emean , (2)

14 where E mean is a mean error of the measurement points in the geometric model.
15 3. Control of accuracy by tying the adjacent models (in the overlapping area or triplets).
16 Following measurements of the tie points on the stereo pairs, the points that belong to the
17 overlapping area of adjacent models (i.e. triplets) were transferred. The relative accuracy of the
18 orientation was verified by comparing the difference of point measurements (triplets) of the
19 adjacent models. Triplet errors E x , E y , E z in 3D coordinate system (X, Y, Z coordinates) were

20 calculated for two adjacent models. The mean of triplet errors in XY plane and Z coordinate were
21 calculated using the following equations 3 and 4:

22
xy
E mean  2  0.5 pxl , (3)
23 where pxl – is the size of the matrix pixel;
c
24
z
E mean   E mean
xy
, (4)
bx
25 where c – is the focal length of the camera, bx is the survey basis on the image scale (mm):
26 bx  l x  100  p x  / 100 , (5)

27 where l x – is the image size along the X-axis (mm), p x – is the size of the overlapping zone (%).
28 The digital professional Canon EOS 1D Mark III model camera equipped with two different
29 lenses with focal length of either 14 mm or 85 mm was used for imaging. The camera was
30 calibrated at the Department of Photogrammetry, Institute of Geodesy and Geoinformation,
31 University of Bonn, Germany, using Tcc software as described previously by Abraham and Hau

9
 1 (1997). The characteristics and the calibration results of Canon EOS 1D Mark III are presented in
2 Tables 1 and 2 (Sužiedelytė-Visockienė and Bručas, 2009).
3

4 Table 1. Characteristics of digital camera Canon EOS 1D Mark III

5
Characteristics Values
Focal length c (mm) 14 and 85
Resolution (pixel) 21 mln.
Pixel pitch (μm) 6.46.4
Image size (mm) 35.923.9
Image size (pixelpixel) 56163744
6
7 Table 2. Canon EOS 1D Mark III calibration results (Department of Photogrammetry, Institute of
8 Geodesy and Geoinformation, University of Bonn)
9
Parameters Results (pixel) Results (pixel)
Focal length
c 2145.98 12928.51
Scale of image (constant)
S xy 0.99 0.99
Correction of the main point of the image
x0 2.41 16.83
y0 -4.093 -9.89
Radial symmetric camera distortion
A1 –9.52E-09 – 5.93E-10
A2 8.29E-16 -3.67E-18
10

11

12 5. Results and discussion

13 To test practicality of methodology, which was developed on the basis of the close-range
14 photogrammetry, an applied study was carried out using digital images of the heritage object.
15 Below we describe key steps of the image acquisition using professional camera and image
16 processing, which were performed on decorative elements (ornaments) in Uzutrakis palace.
17 The images of the chamber wall were taken using the lens with the focal length of 14 mm,
18 while the smaller of the wall containing ornaments were captured with 85 mm lens. Ornaments
19 were stuccoworks, a fine plaster-based artwork often used as three-dimensional ornamentation
20 during a period from 1500 to 1700 (Nardini et al., 2007). The overlapping images of the wall and of
21 the ornaments were corrected to eliminate errors caused by the camera lens distortion reported in
22 Table 2 and are presented in Figures 2, 3 and 4.
23

10
 1
2
3 Fig. 2. Two overlapping images of the wall (c = 14 mm). Ornaments A and B are in dashed circle
4 and dashed rectangle, respectively
5

6
7
8 Fig. 3. Two overlapping images of the ornament (A) (с = 85 mm)
9

10
11
12 Fig. 4. Two overlapping images of the ornament (B) (c = 85 mm)
13

14 In order to perform the interior orientation of images, the central point of images was
15 measured, the coordinates of which were equal to the half size of the image matrix (in pixels).
16 Based on the camera calibration results (Table 2), the position of the point was corrected

11
 1 automatically to eliminate errors of the camera lens distortions. These errors can also be corrected
2 even before starting the image processing (Sužiedelytė Visockiene and Bručas, 2009).
3 Table 3 presents the calculated theoretical mean triple error for Canon EOS 1D Mark III
4 digital camera images with the matrix pixel size of 6 µm, 14 and 85 mm focal length of the camera
5 lens, the image size is 56163744 pixels or 35.923.9 mm, and the images are overlapping by 60,
6 80 and 90%. The calculated error of the horizontal and vertical model is computed from the vertical
7 or horizontal images (Equations 3–5).
8 Often, the relative accuracy of the orientation was assessed using reduced survey basis bx
9 when the overlapping was 60-90%. As a consequence, the ratio of c/bx increased, and therefore, the
z
10 values of triplet errors Emean were also increased (see Table 3).
11

12 Table 3. Acceptable triplets mean error for images

13
Focal length c Overlapping zone px Survey basis bx xy
Emean (µm) z
Emean (µm)
(mm) (%) (mm)
14 60 13.2 4.2 4.5
14 60 8.8 4.2 6.7
14 80 6.6 4.2 8.9
14 80 4.4 4.2 13.4
14 90 3.3 4.2 17.8
14 90 2.2 4.2 26.7
85 60 13.2 4.2 27.0
85 60 8.8 4.2 40.5
85 80 6.6 4.2 54.0
85 80 4.4 4.2 81.1
85 90 3.3 4.2 108.2
85 90 2.2 4.2 162.2
14
15 Following the relative orientation and assessment of triplet accuracy the triangulation
16 adjustment was performed. The acceptable errors for triangulation differ depending on topographic
17 or ortho-photographic models. When compiling topographic maps with the controls points, the
18 adjustment was not allowed to be greater than 0.2 mm in XY plane and 0.15 hint on Z coordinate,
19 where hint contour’s interval of the output map.
20 Acceptable mean of the residuals of tie points was 0.3 mm for the output map scale. The
21 acceptable mean of the residuals of tie points on Z coordinate was 0.2 hint , 0.25 hint , or 0.35 hint ,
22 depending on the contour interval and scale (RACURS, 2012).
23 In PhotoMod system, the acceptable mean of the residuals on CP in XY plane was 0.2 mm
24 on the output map scale and 1/3 hDTM on Z , where hDTM is mean of the residuals of the Digital
25 Terrain Model (DTM). The value of hDTM was calculated using the following equation:

12
 M
1 hDTM  0.3mm  c  , (6)
r
2 where M – is the output map (plane) scale; r – is the maximum distance from the image point to the
3 nadir point (mm), which is equal half of the diagonal of the “working area”. Table 4 presents
4 calculated accuracy requirements for the otho-photographic model.
5

6 Table 4. Parameters representing a quality of ortho-photographic model

7
c (mm) Output plane scale hDTM E XY (µm) E Z (µm)
(mm)
1:100 32.5 20 10.8
14 1:50 16.3 10 5.4
1:10 3.3 2 1.1
1:100 197.7 20 65.9
85 1:50 98.8 10 32.9
1:10 19.8 2 6.6
8
9 In order to evaluate derived values, the ortho-photographic model compiling was assessed.
10 Using equations 1-4, parameters characterising the quality of triangulation were estimated (Table
11 5).
12

13 Table 5. Parameters representing the quality of triangulation

14
Focal Accuracy Accuracy x
E mean (µm) y
E mean (µm) z
E mean (µm) xy
E mean (µm)
length points result (µm)
(mm)
14 CP (10) Emax 2.4 1.8 3.9 2.4
RMS 1.5 0.7 2.1 1.7
Tie (8) Emax 7.4 1.4 - 1.4
RMS 2.1 5.3 - 5.3
85 CP(5) Emax 4.4 5.0 2.3 5.6
RMS 0.7 0.5 1.5 0.9
Tie (4) Emax 0.2 11.5 - 11.5
RMS 0.1 4.3 - 4.3
15
16
17 The comparison of obtained data (Table 5) with the theoretical values in Table 4 revealed
18 that achieved results are of a high accuracy and reliability. It should be noted that due to
19 imprecisions in the approximation of tie points, it was observed greater mean of triplet errors in
20 some cases where tie points were used for triangulation. Altogether, these results allowed to expect
21 positive outcomes form the stereo-digitalization, ortho-photographic model and wall surface data
22 analysis.

13
 1 After stereo digitalizing the topography of the model it was identified that ornaments
2 contained surface and geometric deformations, which were induced by the crack in the surface of
3 the wall and damage/loss of a certain part of the ornament caused by the corrosion. The extent and
4 size of the damaged area was evaluated and thus made it possible to identify the costs and other
5 measures required for the restoration of this architectural heritage. The method of stereo-
6 digitalization was used for compiling structural lines (Figure 5).
7

8
9 Fig. 5. Structural lines of the wall
10

11 The green line was used to mark the main contour of the wall, the red line indicated the
12 surface and geometric deformations of the wall caused by compression and corrosion, and the blue
13 line indicated the contours of the ornaments. The total area of the tested wall was 14.95 m2. The
14 determined area of the deformations amounted to 9.0 m2, which comprised 60 of the total area for
15 the required restoration works.
16 Two ortho-photogarphic models (Figure 6) were compiled to allow drawing of ornaments
17 located on the internal wall by using technique as described previously (Sužiedelytė Visockienė and
18 Bručas, 2010).

19
20

21 Fig. 6. Ortho-photographic models of ornaments A and B

22

14
 1 Using equations 1 and 2 the accuracy of the ornaments of ortho-photographic models were
2 calculated and presented in Table 6.
3

4 Table 6. Parameters representing accuracy of the ortho-photographic model

5

Accuracy E X (µm) EY (µm) E XY (µm)

Ornament A
RMS 2.7 2.9 4.1
Max 4.4 5.0 5.4
Ornament B
RMS 0.7 0.4 0.8
Max 1.1 1.0 1.2
6
7 The obtained ortho-photographic models were of high quality. The accuracy of first
8 ornament met the scale requirement of 1:50 (Table 4). The accuracy of second ornament met the
9 requirement of the scale 1:10. The ortho-photographic models are geometrically correct and
10 coordinated, because of the usage of AutoCad software to be able to draw the structural lines of the
11 ornaments (Figure 7). The surface of the ornaments was not disturbed.
12

13
14

15 Fig. 7. Structural drawings of ornaments A and B

16

17 The structural drawings of ornaments were used for restoration of decorative elements in
18 Uzutrakis palace. They are also stored in the archives of the cultural heritage in Vilnius, Lithuania.
19

20 6. Conclusions
21 The built heritage including the architectural heritage is now recognized not only as an object of
22 cultural value and exclusive property but also as a multidimensional socio-economic asset (Shipley
23 and Reeve, 2010; Gražuleviciute-Vileniske et al., 2011; Hernández-Mogollón et al., 2013; Mark,
24 2013; Rodríguez-Oromendía et al., 2013). As our cultural heritage suffers from various physical and
25 environmental factors and becomes inevitably lost, the documentation and monitoring of heritage

15
 1 objects plays an important role in the architectural heritage preservation. The development of 2D
2 and 3D models of architectural heritage objects in their current state requires methodologies that
3 can digitally preserve and model information about geometry and appearance of such objects for
4 future generations.
5 Here we reported the photogrammetric measurements in conjunction with the exploration
6 studies of the heritage objects to record architecturally valuable elements and their possible surface
7 and geometric deformations. For this purpose, the pipeline of a compilation of the ortho-
8 photographic model for investigation of the heritage object was proposed. With the assistance of
9 appropriate software, such as PhotoMod, it was demonstrated a procedure of entering the area of
10 the geometric deformation onto the ortho-photogtraphic model, which enabled to evaluate the
11 amount of the materials required for the restoration works and to determine the costs required for
12 restoration work.
13 The images of Užutrakis palace walls were used for the analysis. The ortho-photographic
14 model of the wall was compiled by means of the photogrammetric method. The evaluated accuracy
15 was 1-5 µm. Obtained results met the scale requirements of 1:10 and 1:50. The high accuracy of the
16 photographed object was ensured and influenced by the optical quality of the lens of the camera and
17 by the accuracy of the CP and tie points for the object during the photogrammetric measurements.
18 The obtained ortho-photographic model during the photogrammetric measurements was
19 geometrically accurate and orientated. Developed model permitted to depict a precise structural
20 description of the ornaments.
21 The established close-range photogrammetry pipeline can be used for documentation and
22 geometric deformation monitoring of cultural heritage artwork and decorations. Moreover, acquired
23 digital data can be exploited in assessing their state and conditions and help with determination of
24 the quantity of material, which is required for restoration of the ornament. Most importantly, the
25 model, developed utilising the digital data generated by means of the close-range photogrammetry,
26 is easy to store in the digital archive and can be reused for restoration of the architectural heritage
27 object or for other purposes if required in the future.
28

29 Acknowledgements
30 We thank Thomas Cook for reading the manuscript. NM gratefully acknowledges support from the
31 BBSRC (research grants BB/E003192/1 and BB/I008349/1).'
32

33 References
34 Abraham S., Hau T., (1997), Towards autonomous high precision calibration of digital cameras, In:
35 Videometrics V, Proceedings of SPIE, 3174, 82–93.

16
 1 Adrov V.N., Checkurin A.D., Sechin A.Y., Smirnov A.N., Adam-Guillaume J.P., Quessette J.A.,
2 (1995), Program PHOTOMOD: digital photogrammetry and stereoscopic images synthesis on a
3 personal computer. Proceedings of SPIE - The International Society for Optical Engineering,
4 2646, 89-96.
5 Aguilar M.A., Aguilar F.J., Agüera F., (2005), Mapping small areas using a low-cost close range
6 photogrammetric software package with aerial photography. , Photogrammetric Record, 20
7 (112), 335-350.
8 Akman O., Bayramog̃lu N., Aydin Alatan A., Jonker P., (2010), Utilization of spatial information
9 for point cloud segmentation,. 3DTV-CON 2010: The True Vision - Capture, Transmission and
10 Display of 3D Video, art. no. 5506339.
11 Al-kheder S., Al-shawabkeh Y., Haala N., (2009), Developing a documentation system for desert
12 palaces in Jordan using 3D laser scanning and digital photogrammetry, Journal of
13 Archaeological Science, 36(2), 537-546.
14 Arias P., Caamaño J.C., Lorenzo H., Armesto J., (2007), 3D modeling and section properties of
15 ancient irregular timber structures by means of digital photogrammetry, Computer-Aided Civil
16 and Infrastructure Engineering, 22(8), 597-611.
17 Arias P., Herráez J., Lorenzo H., Ordóñez C., (2005), Control of structural problems in cultural
18 heritage monuments using close-range photogrammetry and computer methods, Computers and
19 Structures, 83(21-22), 1754-1766.
20 Arias P., Ordóñez C., Lorenzo H., Herraez J., (2006), Methods for documenting historical agro-
21 industrial buildings: a comparative study and a simple photogrammetric method, Journal of
22 Cultural Heritage, 7(4), 350-354.
23 Barazzetti L., Binda L., Scaioni M. and Taranto P., (2011), Photogrammetric survey of complex
24 geometries with low-cost software: application to the “G1” temple in Myson, Vietnam, Journal
25 of Cultural Heritage, 12(3), 253–262.
26 Barazzetti L., Previtali M., Scaioni M., (2013), Stitching and processing gnomonic projections for
27 close-range photogrammetry, Photogrammetric Engineering and Remote Sensing, 79(6), 573-
28 582.
29 Barazzetti L., Scaioni M., Remondino F., (2010), Orientation and 3D modelling from markerless
30 terrestrial images: Combining accuracy with automation, Photogrammetric Record, 25(132),
31 356-381.
32 Barbetti I., Felici A., Magrini D., Manganelli Del Fa' R., Riminesi C., (2013), Ultra close-range
33 photogrammetry to assess the roughness of the wall painting surfaces after cleaning treatments,
34 International Journal of Conservation Science, 4(SPL.ISS.), 525-534.
35 Beraldin J.-A., Blais F., Boulanger P., Cournoyer L., Domey J., El-Hakim S. F., Godin G., Rioux

17
 1 M., Taylor J., (2000), Real world modelling through high resolution digital 3D imaging of
2 objects and structures, ISPRS Journal of Photogrammetry and Remote Sensing, 55(4), 230–250.
3 Bitelli G., Gatta G., Girelli V.A., Vittuari L., Zanutta A., (2011), Integrated methodologies for the
4 3D survey and the structural monitoring of industrial archaeology: The case of the casalecchio di
5 reno sluice, Italy, International Journal of Geophysics, art. no. 874347.
6 Campana S., Remondino F., (2007), Fast and detailed digital documentation of archaeological
7 excavations and heritage artifacts, Proceedings of 35th Computer Applications and Quantitative
8 Methods in Archaeology Conference, Berlin, Germany, 36–42.
9 Castellini P., Abaskin V., Achimova E., (2007), Portable electronic speckle interferometry device
10 for the damages measurements in veneered wood artworks, Journal of Cultural Heritage, 9(3),
11 225-233.
12 Chandler J. H., Fryer J. G. and Jack A., (2005), Metric capabilities of low-cost digital cameras for
13 close range surface measurement, Photogrammetric Record, 20(109), 12–26.
14 Chandler J.H., Bryan P., Fryer J.G., (2007), The development and application of a simple
15 methodology for recording rock art using consumer-grade digital cameras, Photogrammetric
16 Record, 22(117), 10-21.
17 Chen F., Romice O., (2009), Preserving the cultural identity of Chinese cities in urban design
18 through a typomorphological approach, Urban Design International, 14(1), 36-54.
19 De Reu J., Plets G., Verhoeven G., De Smedt P., Bats M., Cherretté B., De Maeyer W., Deconynck
20 J., Herremans D., Laloo P., Van Meirvenne M., De Clercq W., (2013), Towards a three-
21 dimensional cost-effective registration of the archaeological heritage, Journal of Archaeological
22 Science, 40(2), 1108-1121.
23 Deng G., Falg W., (2001), An evaluation of an off-the-shelf digital close-range photogrammetric
24 software package, Photogrammetric Engineering and Remote Sensing, 67(2), 227-233.
25 Deveikiene V., Deveikis S., (2011), Research of historical and artistic development of the
26 Lithuanian parks created by Edouard André: New facts, discovery, and insights, Town Planning
27 and Architecture, 35(3), 184-199.
28 Fassi F., Achille C., Fregonese L., (2011), Surveying and modelling the main spire of Milan
29 Cathedral using multiple data sources, Photogrammetric Record, 26(136), 462-487.
30 Gražuleviciute-Vileniske I., Janilionis V., Guščinskiene J., Ažukaite L., (2011), Contingent
31 valuation of built heritage properties in a transition country: A case of Lithuania, International
32 Journal of Strategic Property Management, 15(4), 393-415.
33 Gruen A., Akca D., (2005), Least squares 3D surface and curve matching, ISPRS Journal of
34 Photogrammetry and Remote Sensing, 59(3), 151-174.
35 Grussenmeyer P., Landes T., Voegtle T. and Ringle K., (2008), Comparison methods of terrestrial

18
 1 laser scanning, photogrammetry and tacheometry data for recording of cultural heritage
2 buildings, International Archives of Photogrammetry, Remote Sensing and Spatial Information
3 Sciences, 37(B5), 213–218.
4 Guarnieri A., Milan N., Vettore A., (2013), Monitoring of complex structure for structural control
5 using terrestrial laser scanning (Tls) and photogrammetry, International Journal of Architectural
6 Heritage, 7(1), 54-67.
7 Han S.-H., (2012), Digital photogrammetry and laser measurement for preservation and restoration
8 of cultural properties in Korea, Communications in Computer and Information Science, 338, 20-
9 27.
10 Hernández-Mogollón J.-M., Campón-Cerro A.-M., Alves H., (2013), Authenticity in environmental
11 high-quality destinations: A relevant factor for green tourism demand, Environmental
12 Engineering and Management Journal, 12(10), 1961-1970.
13 Hinsch K.D., Gülker G., Helmers H., (2007), Checkup for aging artwork-Optical tools to monitor
14 mechanical behaviour, Optics and Lasers in Engineering, 45(5), 578-588.
15 Kersten T.P., Büyüksalih G., Baz I., Jacobsen K., (2009), Documentation of Istanbul historic
16 peninsula by kinematic terrestrial laser scanning, Photogrammetric Record, 24(126), 122-138.
17 Koutsoudis A., Vidmar B., Ioannakis G., Arnaoutoglou F., Pavlidis G., Chamzas C., (2013), Multi-
18 image 3D reconstruction data evaluation, Journal of Cultural Heritage, 15(1), 73-79.
19 Kutut V., (2011). Technology of heritage objects preservation and management, Technika, Vilnius,
20 Lithuania „Technika“, p. 266.
21 López-Aparicio S., Grašiene R. (2013) Screening indoor air quality evaluation in the Lithuanian
22 theatre, music and cinema museum, Journal of Environmental Engineering and Landscape
23 Management, 21(1), 52-58.
24 LSE, (1983), The Lithuanian Soviet Encyclopedia, Juozapas Tiškevičius IV., vol. XI11, Science
25 and Encyclopedia Publishing Institute, Vilnius: Science and Encyclopedia Publishing Institute, p.
26 640.
27 Martínez S., Ortiz J., Gil M.L., Rego M.T., (2013), Recording complex structures using close range
28 photogrammetry: The cathedral of Santiago de Compostela, Photogrammetric Record, 28(144),
29 375-395.
30 Martorelli M., Pensa C., Speranza D., (2014), Digital Photogrammetry for Documentation of
31 Maritime Heritage, Journal of Maritime Archaeology, 9, 81-93in press.
32 Mazzanti M., (2002), Cultural heritage as multi-dimensional, multi-value and multi-attribute
33 economic good: Toward a new framework for economic analysis and valuation, Journal of
34 Socio-Economics, 31(5), 529-558.

19
 1 Mrak I., (2013), A methodological framework based on the dynamic-evolutionary view of heritage,
2 Sustainability, 5(9), 3992-4023.
3 Nardini I., Zendri E., Biscontin G., Riato S., (2007), Composition and technology of historical
4 stuccoes coming from Grimani Palace in Venice (Italy), Journal of Cultural Heritage, 8(1), 61-
5 64.
6 Nelle A.B., (2009), Museality in the urban context: An investigation of museality and musealisation
7 processes in three Spanish-colonial World Heritage Towns, Urban Design International, 14(3),
8 152-171.
9 Norrström H., (2013), Sustainable and balanced energy efficiency and preservation in our built
10 heritage, Sustainability, 5(6), 2623-2643.
11 Oniga E., Diac M., (2013), Metric and non-metric cameras calibration for the improvement of real-
12 time monitoring process results, Environmental Engineering and Management Journal, 12(4),
13 719-726.
14 Ordóñez C., Arias P., Herráez J., Rodríguez J., Martín M.T., (2008), Two photogrammetric
15 methods for measuring flat elements in buildings under construction, Automation in
16 Construction, 17(5), 517-525
17 Ortiz J., Gil M.L., Martínez S., Rego T., Meijide G., (2013), Three-dimensional modelling of
18 archaeological sites using close-range automatic correlation photogrammetry and low-altitude
19 imagery, Archaeological Prospection, 20(3), 205-217.
20 Pérez-Gracia V., Di Capua D., Caselles O., Rial F., Lorenzo H., González-Drigo R., Armesto J.,
21 (2011), Characterization of a romanesque bridge in Galicia (Spain), International Journal of
22 Architectural Heritage, 5(3), 251-263.
23 Pesci A., Teza G., Bonali E., Casula G., Boschy E., (2013), A laser scanning-based method for fast
24 estimation of seismic-induced building deformation, ISPRS Journal of Photogrammetric and
25 Remote Sensing, 79, 185-198.
26 Pickard R., (2002), A comparative review of policy for the protection of the architectural heritage of
27 Europe, International Journal of Heritage Studies, 8 (4), 349-363.
28 RACURS (2012), Accuracy control at various stages of photogrammetric processing in
29 PHOTOMOD system, Technical support department, Racurs, Moscow, Russia
30 Remondino F., (2011), Heritage recording and 3D modeling with photogrammetry and 3D
31 scanning, Remote Sensing, 3(6), 1104-1138.
32 Remondino F., Girardi S., Rizzi A., Gonzo L., (2009), 3D modeling of complex and detailed
33 cultural heritage using multi-resolution data, Journal of Computing and Cultural Heritage, 2
34 (1), art. no. 2.
35 Remondino F., Rizzi A., (2010), Reality-based 3D documentation of natural and cultural heritage

20
 1 sites—techniques, problems, and examples, Applied geomaticsGeomatics, 2 (3), 85-100.
2 Remondino F., Rizzi A., Barazzetti L., Scaioni M., Fassi F., Brumana R., Pelagotti A., (2011),
3 Review of geometric and radiometric analyses of paintings, Photogrammetric Record, 26(136),
4 439-461.
5 Rodríguez-Oromendía A., Reina-Paz M.-D., Sevilla-Sevilla C. (2013) Environmental awareness of
6 tourists, Environmental Engineering and Management Journal, 12(10), 1941-1946.
7 Sanz J.O., Docampo M.D.L.L.G., Rodríguez S.M., Sanmartín M.T.R., Cameselle G.M., (2010), A
8 simple methodology for recording petroglyphs using low-cost digital image correlation
9 photogrammetry and consumer-grade digital cameras, Journal of Archaeological Science,
10 37(12), 3158-3169.
11 Shipley R., Reeve A., (2010), Evaluating the THI: Measuring the effectiveness of the townscape
12 heritage initiatives in the United Kingdom, Urban Design International, 15(4), 221-237.
13 Solla M., Caamaño J.C., Riveiro B., Arias P., (2012), A novel methodology for the structural
14 assessment of stone arches based on geometric data by integration of photogrammetry and
15 ground-penetrating radar, Engineering Structures, 35, 296-306.
16 Sužiedelytė-Visockienė J., (2013), Accuracy analysis of measuring close-range image points using
17 manual and stereo modes, Geodesy and Cartography, 39(1) 18-22.
18 Sužiedelytė Visockienė J., Bručas D., (2010), The analysis of possibility of creating of terestrial
19 orthophoto, Journal of Vibroengineering, 12(2), 191-199.
20 Sužiedelytė-Visockienė J., Bručas D., (2009), Influence of digital camera errors on the
21 photogrammetric image processing, Geodesy and Cartography, 35(1), 29–33.
22 Sužiedelytė-Visockienė J., Bručas D., Ragauskas U., (2014), Comparison of UAV images
23 processing softwares, Journal of Measurements in Engineering, 2(2), 111-121.
24 Sužiedelyte-Visockiene J., Kumetaitiene A., Bagdžiunaite R., (2011), Accuracy analysis of
25 different surface reconstruction modelling methods for heritage objects digitized considering
26 photogrammetric data, Geodesy and Cartography, 37(2), 56-62.
27 Tapete D., Casagli N., Guido L., Riccadro F., (2013), Integrating radar and laser-based remote
28 sensing techniques for monitoring structural deformation of archaeological heritage. Journal of
29 Archaeological Science, 40(1), 176-189.
30 Tiano P., Tapete D., Matteini M., Cecaroni F., (2008), The Micro-
31 photogrammetry: a new diagnostic tool for on site monitoring of
32 monumental surfaces. International Workshop SMW08, Florence, 27-
33 29 October, 97-106.
34 Varas-Muriel M.J., Fort R., Martínez-Garrido M.I., Zornoza-Indart A., López-Arce P., (2014),
35 Fluctuations in the indoor environment in Spanish rural churches and their effects on heritage

21
 1 conservation: Hygro-thermal and CO2 conditions monitoring, Building and Environment, 82, 97-
2 109.
3 Vergauwen M., Van Gool L., (2006), Web-based 3D reconstruction service, Machine Vision and
4 Applications, 17, 411-426.
5 Yastikli N., (2007), Documentation of cultural heritage using digital photogrammetry and laser
6 scanning, Journal of Cultural Heritage, 8(4), 423-427.
7 Yilmaz H.M., Yakar M., Gulec S.A., Dulgerler O.N., (2007), Importance of digital close-range
8 photogrammetry in documentation of cultural heritage, Journal of Cultural Heritage, 8(4), 428-
9 433.
10 Yilmaz H.M., Yakar M., Yildiz F., (2008), Documentation of historical caravansaries by digital
11 close range photogrammetry, Automation in Construction, 17(4), 489-498.
12 Zheng J., Haiyan G., Yi W., Mingwei S., (2012), 3D reconstruction of small-sized cultural relics
13 based on laser scanning and close-range photogrammetry, International Journal of Digital
14 Content Technology and its Applications, 6(12), 196-205.
15

22