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

net/publication/257515501

Digital speckle pattern interferometry for the condition surveys of painted

wood: Monitoring the altarpiece in the church in Hedalen, Norway

Article  in  Journal of Cultural Heritage · September 2012

DOI: 10.1016/j.culher.2012.01.008

CITATIONS READS

13 150

4 authors:

Łukasz Lasyk Michał Łukomski

Active Text Ltd, Kraków, Poland The J. Paul Getty Trust
15 PUBLICATIONS   70 CITATIONS    64 PUBLICATIONS   491 CITATIONS   

SEE PROFILE SEE PROFILE

Tone marie Olstad Annika Haugen

Norwegian Institute for Cultural Heritage Research (NIKU) Norwegian Institute for Cultural Heritage Research (NIKU)
11 PUBLICATIONS   68 CITATIONS    3 PUBLICATIONS   34 CITATIONS   

SEE PROFILE SEE PROFILE

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

Sustainable Management of heritage Buildings in a Long-term perspective (SyMBoL) View project

Managing Collection Environments Intitiative View project

All content following this page was uploaded by Michał Łukomski on 11 October 2017.

The user has requested enhancement of the downloaded file.

Digital speckle pattern interferometry for the condition surveys of painted

wood: monitoring the altarpiece in the church in Hedalen, Norway

Łukasz Lasyk1, Michał Łukomski1*, Tone Marie Olstad2, Annika Haugen2

1
Jerzy Haber Institute of Catalysis and Surface Chemistry, Polish Academy of

Sciences, ul. Niezapominajek 8, 30-239 Kraków, Poland,

2
The Norwegian Institute for Cultural Heritage Research, Storgata 2 Postboks

736, Sentrum N-0105 Oslo, Norway

* corresponding author: nclukoms@cyf-kr.edu.pl

Abstract

Digital Speckle Pattern Interferometry (DSPI) and Speckle Decorrelation (DIC) were

used in condition surveys of a wooden altarpiece in the church of Hedalen, Norway. Two

surveys were conducted, one before and one after the heating season in the church to

trace the possible development of damage in the paint layer caused by relative humidity

variations induced by the heating system. The measurements demonstrated that the

speckle techniques can contribute to detecting irregular areas on the paint surface and in

1
the paint structure. They proved particularly effective in tracing paint detachments at an

incipient stage which cannot be easily detected by an unaided eye or manual inspection.

The results of the speckle techniques may thus guide a traditional conservation survey, or

the use of further microscopic or analytical survey techniques. The speckle techniques

could be routinely used by conservation practitioners who are not scientists if simple,

portable instruments were available on the market.

Research aims

The main objective of this study was to use Digital Speckle Pattern Interferometry (DSPI)

and Speckle Decorrelation (DIC) in condition surveys of a painted wooden objects, on

site in historic buildings, to trace the development of damage in the paint layer at an

incipient stage which cannot be easily detected by an unaided eye or manual inspection.

Introduction

This paper builds on extensive earlier experience in applying the speckle techniques to l

condition surveys of polychrome wood. It has been the main purpose of the study, which

is reported here, to make a further step on the path to make the speckle techniques simple,

precise and repeatable enough so that they find application for routine condition surveys

executed by non-scientists in situ. Another aim of the study was to establish a link

between the climatic conditions in the church and possible new flaws in the paint layer.

Repeated examinations of mediaeval objects in selected churches in Norway has helped

to establish that the need for conservation and, especially, re-conservation is due to the

fact that the climate in heated churches in cold climate is unfavourable for painted

wooden objects [1-3]. As a result, authorities responsible for the conservation of the

churches have promoted the implementation of localised heating systems. The idea has

been to limit the heating to the areas where people are, while changing the conditions in a

2
church as little as possible, thus favouring the conservation of its historic furnishings [4-

6].

The object of the survey was a medieval altarpiece in the periodically heated stave church

in Hedalen, Norway.

1. Condition surveys in general

Condition surveys are day-to-day occupation of conservators. They are a principal tool in

getting insights into causes of damage to objects, essential for improving management

and guiding research. Particularly important are conditions surveys or examinations

repeated at regular time intervals to assess a dynamic damage process. A condition survey

of a painted surface documenting and mapping failures and damage areas in the paint

structure is a crucial step before a treatment so that the latter can be adequately discussed,

planned and justified. The most important survey tools are conservator’s eyes, experience

and skill. In many institutions, forms exist that guide the conservator through the

examination of the object. Looking for loose paint, which is an important part of the

survey and the part we will concentrate on here, is done in many ways. The use of optical

microscopy, raking light and tapping on the surface are the methods most often used. The

method of using the sound feedback is named the acoustic or percussion method [7]. The

surface of the painting is gently tapped with a nail or a finger or a suitable lightweight

instrument, like a small stick. The acoustic response to the tapping gives to the trained ear

a general indication of irregularities in the paint layer, like loose paint. Often a suspicion

that there is loose paint has to be confirmed by optical microscopy. Usually, a more high-

pitched sound indicates a compact and sound structure while a low-pitched sound is

evidence of hollows and indicates a loose layer. The degree of deformability of detached

areas is tested by exerting a slight pressure. The method is not free of risks because

3
mechanical pressure is used on a possibly fragile structure; but a trained eye and hand

avoids loss when examining the surface. The person examining the surface has to survey

the total surface with the same level of concentration. There is always a possibility that

some loose paint may ‘escape’ conservator’s attention, and microscopic flaws may not be

detected.

Another weak point of the acoustic methods is that they do not quantify the amount of

loose paint. Usually, the conservator gives a qualitative description of the condition and,

for the documentation, the areas of loose paint are often marked on a photograph.

Also when more sophisticated methods are employed, like active infrared thermography,

a conservator has to interpret the results to explain the condition of the paint.

An aim is therefore to supplement the photographic documentation and the conservator’s

traditional visual observation with scientific methods for tracing flaws like loose paint.

Interferometric techniques have been increasingly used as diagnostic tools, as they are

non-invasive and because they provide detailed, full-field information on the surface of

the object. However, as the information is restricted to the inspected area, it has to be

transferred into a general description of the condition for the total paint surface. To obtain

good understanding of damage occurring within a collection in general, or more

specifically, in the paint layer on a wooden object, the surveys should provide

quantitative information for the same object at different points of time, and the current

damage, should be clearly distinguished from the damage caused in the past.

2. Surveys and speckle techniques

Digital speckle pattern interferometry (DSPI) and speckle decorrelation (DIC) are

particularly attractive for investigations of painted wood. The DSPI technique makes use

of speckles – a granular pattern of light and dark – produced whenever a rough surface is

4
illuminated by a laser light. The equipment needed is portable and the method is capable

of mapping paint layer displacements to a fraction of a micrometer. Damage areas at the

micro-level can be traced before they are discernible visually. The successful applications

have resulted both from technical developments – exemplified by the use of fiber optics

in the construction of the interferometers [8] – and improvements in the image processing

algorithms [9]. To the authors’ knowledge, DSPI was used for the first time in an onsite

work in 1993 to map damage areas in wall paintings [10]. Thorough description of

advantages and limitation of interferometric techniques applied for the analysis of panel

paintings can be found in a review paper by Ambrosini and Paoletti [11]. Both DSPI and

DIC techniques have been applied to examinations of panel paintings [12-14] and other

wooden artworks [15-16]. Recent development of lasers as well as computerized,

automatic fringe recording and processing algorithms make laser speckle metrology an

effective, non-destructive tool with significant advantages compared to alternative

monitoring techniques [17-19].

The described developments have opened the perspective of a routine application of the

speckle techniques for examinations of painted wood, though reducing the cost of the

equipment, and simplifying the measuring procedures and interpretation of the results

remain crucial tasks.

3. Condition survey in the Hedalen stave church

The need to carry out the condition surveys after the conservation in 2007 was dictated by

a concern that instable indoor climate in the church with episodes of severe drops in the

relative humidity due to heating in the cold period, may adversely affect the polychrome

wood. Incidents of serious climate induced damage to the polychrome sculptures

preserved in Hedalen church were recorded in the past [1].

5
The condition surveys using the speckle techniques were carried out twice: in mid-

November 2008 and then in April 2009 i.e. before and after the coldest season during

which the heating system was operated. The surveys were done on selected parts of the

altarpiece in the church.

3.1. Climatic conditions in the Hedalen stave church

The church is a wooden construction. The western part of the nave which is the original

twelfth century stave church is not insulated, the part added in the eighteenth century and

restored in mid-twentieth century, the eastern part of the nave and the chancel is partly

insulated. In the summer, the climate inside the unheated church practically follows the

conditions outdoors.

During the cold season, the church has an intermittent heating with the continuous

background heating set to maintain a minimum temperature of 7 oC. During the

monitored period, the church was heated to 18-19oC, when used. The church has been

heated since 1903; with ovens under the benches since 1957. The climate in the church

was monitored in the periods 1986-1994 (at selected time intervals), 2001-2002, 2002-

2003 and again in 2008-2009. The climatic conditions in the church are better than

expected in an intermittently heated wooden church in inland Norway, but the difference

between the actual climatic conditions and the ideal conservation conditions is huge. The

heating episodes cause periodic fluctuations of RH when it drops first from high to low

levels and then returns to high RH after the heating is reduced. The maximum RH

fluctuation recorded in the winter of 2008/2009 was 38% over a period of 36 hours,

whereas absolute minimum and maximum values of RH recorded in the church during

the monitoring period were 21% and 72% respectively. In general 10-15% fall in RH is

recorded when the church is heated.

6
 3.2. The altarpiece in Hedalen church

The altarpiece is a mediaeval wooden polychrome crucifix mounted on a background.

The wooden (oak), painted background: a central part with two wings on each side, was

originally a tabernacle for a thirteenth century Maria sculpture. In the eighteenth century,

the tabernacle was separated from the sculpture, redecorated and placed on the altar as the

background for the redecorated medieval crucifix [20]. The altarpiece still has much of

the original mediaeval decorative layer beneath the eighteenth century paint. The area for

the combined surveys was selected on one of the wings. Originally the wings were

decorated with small sculptures placed against a golden background. Therefore, beneath

the eighteenth century polychrome decorative scheme on the wings, there is mainly a

layer of resin varnished silver leaves on top of a layer of bole on a chalk-glue ground. The

metal leaf areas have incisions both in the ground and in the layer of metal.

The binding medium of the eighteenth century paint is oil. The thickness of the paint

layer varies over the surface; both because of various colours being superimposed one

over another and the amount of paint in the brushstrokes. The tabernacle was cleaned and

conserved in 1974, before it was last conserved in 2007 [20].

3.3. Digital speckle interferometry (DSPI) used on the altarpiece in the

church

The left bottom panel of the right wing of the altarpiece, marked in Figure 1, was selected

for the analysis. The selected part of the panel was divided into 44 fields measuring 25 x

30 mm which were individually analysed by DSPI. Figure 2 shows the size of the field

illuminated by a laser beam during the measurement. The relatively small size of the

fields analysed increased image quality and produced fringe patterns of good resolution,

7
in the range of millimetres. The images optimally revealing various features of the

painted surface were combined digitally, so that they could be compared with

photographs of the panel.

The DSPI survey was carried out independently of a traditional condition survey.

A schematic diagram of the optical setup to perform DSPI analysis used in this study is

shown in Figure 3. A laser beam from a continuous-work diode-pumped Nd:Yag laser

(100 mW) from Power Technology is split into two beams using a glass plate. These are

referred to as the object and reference beams. The object beam illuminates the

investigated surface while the reference beam passes through the ground glass and is

merged with the beam reflected from the investigated object using a beam-splitter cube.

The object and reference beams superimpose coherently producing an interferogram

recorded by a BCi4-6600 CCD camera from C-cam Technologies (6.6 MPixels with 3.5 x

3.5 pixel pitch). The interferometric filter, centred at the laser wavelength, filters out

optical noise and makes possible daytime operation. When a deformation of the same

order as the wavelength of the laser is induced on the investigated surface, a change in the

speckle pattern occurs.

In this study, thermal- and sound-induced displacement of the object’s surface was used.

In the first approach, the surface is heated with a flow of warm air for a few seconds,

resulting in a recorded increase of the surface temperature of approximately 2-3oC.

Immediately after the heating, the speckle pattern on the surface is recorded as a reference

image, then the consecutive images are subtracted arithmetically using the computer and

the difference is displayed on the monitor. It has a form of a series of closely spaced light

and dark bands known as fringes. The fringe pattern is formed as a result of different

local out-of-plane displacements related to different rates of cooling across the surface.

For example, a detached paint layer cools differently from the firmly attached one, as a

8
thick paint accumulation does when compared to a thin paint layer. Generally, fringes

become clustered and distorted around local surface deformations and may be used as a

guide to the identification of any heterogeneities in the paint layer. The measuring

procedure allows monitoring the formation of fringes in real-time, which gives the

operator a possibility of selecting and ‘freezing’ a suitable record clearly displaying a

feature so that its size and shape are optimally defined. Generally, the patterns were

selected for storage in the computer memory at time intervals of approximately one

minute. The described method was particularly well suited to the analysis of planar

features of the paint layer. By way of example, Figure 4 illustrates fringe patterns due to a

retouched loss in the paint and a glued-back detachment in the paint layer next to the loss.

3.4. Sound-induced excitation of the object’s surface

Relating fringe patterns to defects, especially paint layer detachments, remains difficult.

DSPI measurements were therefore repeated using sound-induced excitation of the

surface for the areas in which paint detachments were suspected. It causes vibration of the

detached paint when the sound wave is close to its resonant frequency. This vibration

produces islands of fringes distinguishing the detached areas from the well adhered paint.

A sound signal with controlled frequency and amplitude was generated by a National

Instrument NI PCI 6221 computer card, amplified and transformed to a sound wave by a

Monacor MPT – 177 loudspeaker with a frequency range from 3.5 to 20 kHz. A

quantitative analysis of the vibration amplitude of the detached paint requires a precise

control of the phase relation between the object and reference beams. This control is

achieved by installing a liquid crystal phase shifter (ARCoptix) in the object beam path.

The measuring procedure consists of scanning sound frequencies during consecutive

measurements to find resonant frequencies (characteristic) of all delaminated areas. When

9
the resonant frequency is established, a phase-stepping measurement is conducted and the

consecutive analysis provides precise information on the characteristic frequency of the

delaminated area, its size and the spatial distribution of the vibration amplitude. Readers

interested in further technical details are referred to [21]

3.5. Speckle decorrelation

Speckle decorrelation measurements were performed using the same DSPI equipment

(Fig. 3) in which the reference beam was blocked. Immediately after switching off the

surface heating, the speckle pattern on the illuminated surface was recorded as a reference

image, and then the consecutive images recorded at time intervals below one minute were

subtracted arithmetically from the reference one, and displayed on the computer monitor.

If the images were perfectly correlated they would cancel completely when subtracted. If

there was some difference - in our case different deformation of irregular areas on cooling

– the irregularities were visible as bright areas. The method was particularly well suited to

recording hair-like fractures in the decorative layer. By way of example, Figure 5 shows

cracking patterns in the paint layer as revealed by the method.

3.6. Traditional condition survey used on the altarpiece the church

A traditional condition survey based on visual observation with the help of raking light

and magnifiers, as well as acoustic analysis: careful tapping of the surface of the paint

layer using a small stick, was carried out on the selected area of the wing of the altarpiece.

Two paintings conservators conducted surveys independently of each other.

The procedure was as follows: the first conservator checked the areas with irregularities

detected with DSPI to see if they were caused by loose paint. A list of comments was

10
made. The second conservator examined the selected area of the panel without knowing

the results of the first survey.

Then the results of the two surveys were compared and the conservators inspected again

areas of the painting which produced fringe anomalies, especially those in which sound

excitation had induced vibration.

4. Results and discussion

The DSPI technique allowed determination of expansion of the irregularities in the paint

layer by comparing the contours of respective areas recorded by the two consecutive

surveys before and after the heating season in the church, as illustrated in Figure 6.

DSPI detects all irregularities in the surfaces and a paintings conservator needs to discern

between flaws that should be treated or monitored, and mere irregularities in the paint

layer.

Figure 7 compares the outcome of the condition surveys. The DSPI survey revealed 15

areas with irregularities. These are marked in the figure with circles. Only 3 of these were

defined by the conservators to represent loose paint. These are marked with rectangles.

The fringe anomalies for the other 12 were found to be due to lost paint, discontinuities in

the surface between retouching and the original paint, cuppings, or irregularities caused

by the painting technique, like incisions in the original ground. Examples of features

other than loose paint, and the corresponding fringe patterns are shown in Figures 4 and 8.

Some of the flaws detected by DSPI, were so minor that the paintings conservator hardly

would have seen them without the guidance of the DSPI. The areas of loose paint

detected in the tabernacle were too small to be treated, which again shows the advantage

of using DSPI to detect areas where the paint starts to lose its adhesion. Such areas may

be monitored and eventually treated before the paint falls off.

11
The sound-induced DSPI technique seems to be a better tool for discriminating between

irregularities and flaws than the heat-induced DSPI. The latter technique detected one

detachment of the paint layer which could not have been easily seen during a first survey

with the unaided eye: an L- shaped detached area about 5 mm high and 2 mm wide is

shown in Figure 6. The area is revealed by a cluster of distorted fringes concentrated

along a contour of the detachment. But the precise extent and shape of the detached paint

was only established by measuring sound-induced vibration of the surface, as illustrated

in the same figure. Such analysis is only possible when the detached paint flake is big

enough to vibrate with the amplitude comparable to the laser wavelength.

The described investigations gave also evidence that the climatic conditions in the church

did not practically produce any new flaws in the paint layer over the time interval

between the two DSPI surveys during which the heating system operated. The survey of

the paint condition using DSPI should preferably be repeated at regular time intervals as

damage development depends not only on the RH fluctuation amplitude but also on

number of the fluctuations. Therefore, the harmlessness or otherwise of the existing

climatic conditions can be confirmed over more heating seasons only.

5. Conclusions

The study has shown that laser speckle techniques have considerable potential to facilitate

and render more precise in situ condition surveys of painted wood. The main findings are

as follows:

- DSPI can greatly contribute to monitoring planar features of the paint layer. By

recording anomalous fringe clusters, a map of surface irregularities can be

12
 produced and guide a subsequent traditional conservation survey using visual

inspection, or further microscopic or analytical techniques

- speckle decorrelation technique in turn is particularly useful in tracing the

fracture development in the paint layer

- DSPI may reveal paint detachments at an incipient stage, which cannot be easily

detected by visual or manual inspection

- the speckle techniques are precise and repeatable enough to follow the defect

development between subsequent condition surveys

- the techniques can be attractive tools for conservators who are not-scientists.

When simple, portable instruments become available on the market, laser spackle

techniques will become a tool for the conservator in checking conservation work, guiding

surveys and hopefully in quantifying the area of loose paint.

Acknowledgments

The research was supported by grant PL0086 from Iceland, Liechtenstein and Norway

through the European Economy Area Financial Mechanism. Michał Łukomski gratefully

acknowledges further support received from the Polish Ministry of Science and Higher

Education, grant POL-POSTDOC II PBZ/MEiN/01/2006/10, and COST Action IE0601

‘Wood science for cultural heritage’. The work on site was made possible by the

Directorate for Cultural Heritage in Norway, and by the local Church Authorities. Mille

Stein, a paintings conservator/researcher at the Norwegian Institute for Cultural Heritage

Research, collaborated in the church with the project team. Furthermore, presentations

and discussions of various parts of this research during meetings of the IE0601

COST Action have significantly contributed to the progress of this study.

13
References

1. Marstein, N., and Stein, M., Advanced measuring of the climatic conditions in the

medieval wooden churches in Norway, in: ICOM Committee for Conservation, 8th

Triennial Meeting, Sydney, Australia, 6-11 September 1987: Preprints, ed. K.

Grimstad, Getty Conservation Institute, Los Angeles (1987) 889-96.

2. Olstad, T.M., Mediaeval wooden churches in cold climate – parish churches or

museums?, in: Preventive Conservation: Practice, Theory and Research, ed. A.

Roy and P. Smith, International Institute for Conservation, London (1994) 99-103.

3. Olstad, T.M., and Stein, M., Saving Art by saving energy, NIKU Temahefte (2),

Norwegian Institute for Cultural Heritage Research (1996).

4. Olstad, T.M., and Haugen, A., Klima i stavkirker: Lokal klimatisering av

menigheten i Kaupanger stavkirke, Sogndal kommune, NIKU Fagrapport (4),

Norwegian Institute for Cultural Heritage Research (1997) [in Norwegian].

5. Bratasz, L., Kozlowski, R., Camuffo, D., and Pagan, E., Impact of indoor heating

on painted wood: monitoring the mediaeval altar in the church of Santa Maria

Maddalena in Rocca Pietore, Italy, Studies in Conservation 50 (2007) 199-210.

6. Camuffo, D., Pagan, E., Rissanen, S., Bratasz, L., Kozlowski, R., Camuffo, M.,

and della Valle, A., An advanced church heating system favourable to artworks: A

contribution to European standardisation, Journal of Cultural Heritage (2009)

doi:10.1016/j.culher.2009.02.008.

7. Hinsch, K.D., Zehnder, K., Joost, H., Gülker,G., Monitoring detached murals in the

Convent of Müstair (Switzerland) by optical metrology, Journal of Cultural

Heritage 10 (2009) 94-105.

14
8. Paoletti, D., Spagnolo, G.S., Facchini, G.M., and Zanetta, P., Artwork diagnostics

with fiber-optic digital speckle pattern interferometry, Applied Optics 32 (1993)

6236-6241.

9. Huntley, J. M., Automated Analysis of Speckle Interferograms, in: Digital Speckle

Pattern Interferometry & Related Techniques, ed. P.K. Rastogi, Wiley, New York

(2000) 59–139.

10. Paoletti, D., and Spagnolo, G.S., Application of fibre optic digital speckle

interferometry to mural painting diagnostics, Measurement Science and

Technology 4 (1993) 614-618.

11. Ambrosini, D., and Paoletti, D., Holographic and speckle methods for the analysis

of panel paintings. Developments since the early 1970s, Reviews in Conservation 5

(2004) 38-48.

12. Lucia, A.C., Zanetta, P.M., and Facchini, M., Electronic speckle pattern

interferometry applied to study and conservation of paintings, Optics and Lasers in

Engineering 26 (1997) 221-233.

13. Albrecht, D., Franchi, M., Lucia, A.C., Zanetta, P.M., Aldovrandi, A., Cianfanelli,

T., Riitano, P., Sartiani, O., and Emmony, D.C., Diagnostic of the conservation

state of antique Italian paintings on panel carried out at the Laboratorio di Restauro

dell’Opificio delle Pietre Dure in Florence, Italy with ESPI-based portable

instrumentation, Journal of Cultural Heritage 1 (2000) 331–335.

14. Tornari, V., Optical and digital holographic interferometry applied in art

conservation diagnosis, e-Preservation Science 3 (2006) 51-57.

15. Spagnolo, G.S., Ambrosini, D., and Paoletti, D., Image decorrelation for in situ

diagnostics of wooden artifacts, Applied Optics 36 (1997) 8358–8362.

15
16. Castellini, P., Abaskin, V., and Achimova, E., Portable electronic speckle

interferometry device for the damages measurements in veneered wood artworks,

Journal of Cultural Heritage 9 (2008) 225- 233.

17. Hinsch, K.D., Laser speckle metrology - a tool serving the conservation of cultural

heritage, in: Oscillation, Waves and Interaction, ed. T. Kurz, U. Parlitz and U.

Kaatze, Universitätsverlag Göttingen (2007) 259-278.

18. Hinsch, K.D., Gülker, G., and Helmers, H., Checkup for aging artwork—Optical

tools to monitor mechanical behaviour, Optics and Lasers in Engineering 45 (2007)

578–588.

19. Bernikola, E., Nevin, A., and Tornari, V., Rapid initial dimensional changes in

wooden panel paintings due to simulated climate-induced alterations monitored by

digital coherent out-of-plane interferometry, Applied Physics A 95 (2009) 387–

399.

20. Stein, M., A144 Hedalen stavkirke, Sør Aurdal kommune, Oppland. Undersøkelse

og behandling av madonnaskapets korpus. NIKU rapport 65/2, Norwegian Institute

for Cultural Heritage Research (2008) [in Norwegian].

21. Lasyk, L., Lukomski, M., and Bratasz, L., Simple DSPI for the investigation of art

objects, Optica Applicata 41 (3) 2011, 687 - 700.

16
Caption for figures

1. Main altar of the church in Hedalen, Norway. The conditions surveys were

performed for the painted wooden panel marked.

2. DSPI system on the altar during the measurements - the field illuminated by a laser

beam is visible.

3. Schematic of the optical setup to perform DSPI analysis. During speckle

decorrelation measurements reference beam is not used.

4. Fringe pattern showing an area of thick paint.

5. Detection of cracking pattern in the paint layer by speckle decorrelation.

6. A detachment of the paint layer at an incipient stage: (a) photograph of the

investigated area; (b) a contour of the detachment revealed by a cluster of distorted

fringes; (c) the analysis of sound-induced vibration of the surface; (d) the same

defect was much reduced in size during the DSPI survey carried out 5 months earlier.

7. Irregular or problematic areas on the painting as detected by a traditional

conservation survey and by DSPI, marked by rectangles and ovals, respectively.

8. Fringe pattern showing the presence of thick paint (top) and discontinuity between

retouching and the original paint (bottom).

17
Fig.1

Fig. 2

18
Fig. 3

Fig. 4

Fig. 5

19
Fig. 6

Fig. 7

20
 Fig. 8

21

View publication stats