ANALYTICAL IMAGING OF CULTURAL HERITAGE BY

SYNCHROTRON RADIATION AND VISIBLE LIGHT- NEAR

INFRARED SPECTROSCOPY

Jay Arre Toque, Yuji Sakatoku, Julia Anders, Yusuke Murayama and Ari Ide-Ektessabi

Advanced Imaging Technology Laboratory, Graudate School of Engineering

Kyoto Univeristy, Yoshida-honmachi, Sakyo-ku, 606-8501, Kyoto, Japan

Keywords: Analytical imaging, Spectral reflectance, Synchrotron radiation, Visible light, Near infrared, Multispectral

imaging, Cultural heritage.

Abstract: Imaging is an important tool for analyzing cultural heritage. Due to its delicate nature, the analysis presents

numerous technical challenges, probably the most important of which is its requirement for non-destructive

and non-invasive investigation. In this study, two techniques used in the analysis of cultural heritage are

presented. The first one, synchrotron radiation x-ray fluorescence, is an advanced analytical technique with

high accuracy and good spatial resolution. On the other hand, spectroscopic technique based on visible

light-near infrared spectrum is becoming popular due to some information that it can provide, which are not

available even in advanced analytical techniques. These two techniques were used to analyze real cultural

heritage such as an ancient Mongolian textile, traditional Korean painting and commonly used pigments in

Japanese paintings. The results revealed that using synchrotron radiation-based techniques is sometimes not

enough in providing critical information (e.g. spectral reflectance, color, etc.) necessary for understanding

of cultural heritage. This can be complemented using visible light-near infrared technique.

1 INTRODUCTION

Cultural heritage refers to artifacts and intangible

features inherited from previous generations, which

are preserved or maintained for the benefit of future

generations. The cultural heritage which we

inherited, show the techniques, the way of life, the

social values and the way people were thinking in

old days. In the past, interests in cultural heritage

are mainly based on its artistic and historic values.

However during the recent years, it has been

attracting the attention of scientists and engineers

because of the technical challenges it presents during

analysis, restoration and preservation. Its delicate

nature requires that the investigation should be non-

destructive and non-invasive (Faubel, et al., 2007;

Toque, et al., 2008).

Among the available analytical techniques,

synchrotron radiation-based analysis offers high

precision and high accuracy in addition to being

non-destructive. Synchrotron radiation (SR) is

emitted in the tangential direction when electrons

and positrons are accelerated at relativistic speed

subjected to a magnetic field (Margaritondo, 1988).

The energy of SR covers a broad spectrum. X-ray

fluorescence analysis using synchrotron radiation is

a powerful technique in terms of detecting ultra trace

elements and studying them in detail. Its unique

features include local area analysis by using micro

beam; capability of doing measurement in air or

water; non-contact and non-destructive assay; rapid

measurements; and precise assay of trace elements

(Ide-Ektessabi, 2007). However, even though SR-

based techniques are powerful, there are information

they cannot provide, such as spectral reflectance,

color information and others. These are important in

analyzing cultural heritage.

To compensate for some inadequacies of SR-

based techniques, imaging at the visible light-near

infrared (VL-NI) spectrum is employed. The basic

idea is to use polychromatic images (with RGB

tristimulus values) or multispectral images to extract

spectroscopic data. It has been reported that

materials emit specific spectral features within a

121

Toque J., Sakatoku Y., Anders J., Murayama Y. and Ide-Ektessabi A. (2009).

ANALYTICAL IMAGING OF CULTURAL HERITAGE BY SYNCHROTRON RADIATION AND VISIBLE LIGHT- NEAR INFRARED SPECTROSCOPY.

In Proceedings of the First International Conference on Computer Imaging Theory and Applications, pages 121-128

DOI: 10.5220/0001788201210128

 SciTePress

certain range (Balas, et al., 2003). In other

wavelength range, this is more pronounced.

However, at the VL-NI range the interaction is more

complex. Nonetheless, there are some unique

features that are only observable within this range.

This makes it valuable in cultural heritage analysis.

2 EXPERIMENT

2.1 Imaging and Spectroscopic

Synchrotron Radiation Analysis

Synchrotron radiation X-ray fluorescence (SRXRF)

was performed on a 13

century Mongolian textile

and dislodged fragments collected from an old

Korean painting. Measurements were done at beam

line 4A of Photon Factory. The electron beam

energy in the storage ring was 2.5 GeV, with a

maximum current of 400 mA. Incident X-ray

energy was 15 keV. The cross-section of the beam

was approximately 1(v) x 1(h) mm

on the sample.

The synchrotron radiation was monochromated by a

multilayered reflecting mirror. Precise beam size of

monochromated X-rays was adjusted using slits. The

incident and transmitted X-rays were monitored by

ionization chambers that were set in front of and

behind the sample. The fluorescent X-rays were

collected by a solid-state detector at 90 degrees to

the incident beam. Measurements were performed in

air. Point spectra were measured for obtaining

consistent elements of the samples. The spectra were

obtained by using a multi-channel analyzer. The

measurement time was 100 seconds for each

spectrum. XRF imaging technique was applied in

order to investigate the distributions of main

elements. X-Y step pulse motors moved the sample

stage. The measurement areas were divided into

matrices of 20 x 20 pixels. At each pixel, the XRF

yields for each element were integrated by single

channel analyzers. The measurement time was three

seconds for each pixel.

Synchrotron radiation was also used to

investigate the relationship between fine structural

change and color fading in natural mineral,

specifically azurite and malachite. Heating of natural

pigments is well practiced among traditional

Japanese painters to modify the shade and color of

the pigments. It is of great interest to understand the

factors involved in this process. Ten samples of each

pigment were heated at 260°C with holding time

from 10 minutes up to 90 minutes at 10-minute

increments. The spectral reflectance of the heated

and unheated pigments was measured to track the

changes in color. X-ray fluorescence and X-ray

absorption fine structure (XAFS) were used to

characterize the pigments. The incident X-ray was

15 keV for XRF analysis while the energy was

scanned at the Cu K absorption edge from 8.90 to

9.09 keV for XAFS.

2.2 Imaging at the Visible-Near

Infrared Range

In order to use VL-NIS for the analytical imaging of

cultural heritage, a technique for image analysis was

developed. Figure 1 illustrates the basic scheme of

the analysis. It begins by capturing an image of the

object to capture spectral and color information. The

picture can be an RGB or a multispectral image. An

RGB image refers to an image captured using

tristimulus values corresponding to red, green and

blue colors. A multispectral image, on the other

hand, is an image captured by using 6-7 bands of

color and infrared filters. The images are captured

within a certain wavelength band. The color

information is then used to simulate the geometrical

and light flux conditions. The simulation enables the

estimation of the spectral reflectance, which is used

for comparison with a database to provide useful

information about the image. The database includes

more than 1000 mineral pigments and is

continuously being updated. The database also

includes additional data such as SRXRF spectra,

XRD spectra as well as information about the

artwork’s history.

Figure 1: Image analysis scheme using VL-NIS.

IMAGAPP 2009 - International Conference on Imaging Theory and Applications

122

Figure 2: (a) SRXRF spectra of an ancient Mongolian textile. Sample A refers to the gold thread while Sample B refers to

the textile; (b) Field-emission SEM showing textile structure.

Figure 3: (a) Optical image of an ancient Mongolian textile laced with gold thread; (b) Elemental distribution image of gold

derived from the SRXRF spectra.

3 RESULTS AND DISCUSSION

3.1 Synchrotron Radiation of Actual

Cultural Heritage

3.1.1 Ancient Mongolian Textile

SRXRF was used to detect the constituent elements

used for the manufacturing of the ancient textile as

well as obtaining elemental distribution images.

Figure 2a shows the XRF spectra of the Mongolian

textile. The spectra revealed the presence of

significant amount of gold and iron. The textile was

dated to be about 700-800 years old. It is believed

that the textile was produced during the 13

century

at the height of power of the Mongolian empire

under Genghis Khan and Kublai Khan. This may

explain why gold was detected from the XRF

spectra. Lacing and decorating the textile with gold

was a symbol of wealth and prosperity in the old

days. The spectra also show traces of Cu and Ti.

These elements are well-known metallic mordant

and are believe to be widely used during that period.

Fig. 2b shows a field-emission SEM image of the

Mongolian textile sample. It depicts textile structure

and weaving condition. In addition, Figure 3a shows

an optical image of a textile laced with gold thread.

An elemental image distribution of gold (Figure 3b)

is derived from the XRF spectra.

3.1.2 Old Korean Painting

SRXRF was used to investigate a traditional Korean

painting. Figure 4 shows the spectra of several

natural pigments used (white, blue, green and red).

The main elements detected were Pb, Cu, Hg and

Fe. It is interesting to note that the spectra of the

blue and green pigments are very similar to popular

pigments used in traditional Japanese painting

known as gunjo (azurite) and ryokusho

(malachite). These pigments are copper-based

pigments [2CuCO

•Cu(OH)

and CuCO

•Cu(OH)

respectively]. The difference lies with the significant

trace of Pb found in the Korean painting. It is

ANALYTICAL IMAGING OF CULTURAL HERITAGE BY SYNCHROTRON RADIATION AND VISIBLE LIGHT-

NEAR INFRARED SPECTROSCOPY

123

Figure 4: SRXRF of different pigments used in a

traditional Korean painting. Note: (a) white pigment; (b)

blue pigment; (c) green pigment; (d) red pigment.

Figure 5: Optical images of the pigments used on a

traditional Korean painting: (a) white pigment; (b) blue

pigment; (c) green pigment; (d) red pigment.

IMAGAPP 2009 - International Conference on Imaging Theory and Applications

124

attributed to the white pigment used in the painting.

It is believed that the white pigment was used as a

base medium. This may explain why all the

pigments studied contain significant traces of Pb.

Figure 5 shows the optical images of the pigments

used on the Korean painting investigated. It also

shows that the pigments used are granulated.

3.1.3 Effect of Heating on Pigment

Discoloration

Figure 6 shows the spectral reflectance of azurite

and malachite pigments as a function of heating

time. These two pigments were selected as test

samples because they are the most commonly used

blue and green pigments in traditional Japanese

paintings. The pigments predictably changed its

color upon heating. It is interesting to note however

that the discoloration did not change the position of

the spectral peak. This implies that the color

wavelength did not change, only the luminosity

(Toque, et al., 2008). The samples got darker with

the increase in heating time. In addition, abrupt

change in color was observed between 40-60

minutes of exposure to elevated temperature.

Figure 6: Spectral reflectance of (a) azurite and (b)

malachite as a function of heating time. Prolonged

exposure to elevated temperature lowers the reflectance of

the pigments.

In order to understand the mechanism of

discoloration of the pigments when subjected to

elevated temperature, they were subjected to

synchrotron XRF and XAFS analysis. Figure 7

shows the XRF spectra of azurite and malachite. It

shows minimal change in the spectra. This implies

that the color change was not due to the trace

elements of the pigments. The trace elements of both

the pigments are similar but their colors are quite

different. The spectral reflectance of azurite leans

toward the short-wavelength range giving it a shade

of blue while malachite is around the mid-

wavelength range giving at a shade of green.

Figure 7: XRF spectra of the heated and non-heated: (a)

azurite and (b) malachite. Both the pigments have similar

spectra. In addition, minimal change is observable

between the burnt and unburned pigments.

The changes in color of both the pigments may

be more attributed to the change in chemical

bonding state of the main element Cu. Figure 8

shows the XAFS spectra of the pigments. It was

found that the absorption edge of the samples heated

for 80 and 90 minutes shifted to lower energy level.

This may also explain the abrupt change in color as

the pigments were heated for a long time.

ANALYTICAL IMAGING OF CULTURAL HERITAGE BY SYNCHROTRON RADIATION AND VISIBLE LIGHT-

NEAR INFRARED SPECTROSCOPY

125

Figure 8: Position of Cu K absorption edge of heat-

discolored (a) azurite and (b) malachite. Unlike the XRF

spectra, XAFS provide more insights about the

discoloration of the pigments.

3.2 Visible Light-Near Infrared

Spectroscopy of Cultural Heritage

Visible light- near infrared spectroscopy (VL-NIS)

gives information about an image, which can

provide better understanding of the mechanism of

degradation and useful insights for the restoration

and preservation of cultural heritage. Some

information about a material can be extracted by

analyzing the interaction between matter and light at

the visible to near infrared radiation spectrum, which

are not noticeable in other range.

The analysis covers the electromagnetic radiation

wavelength from 380-850 nm. This range includes

the visible to near infrared range. IR is included

since recent investigations have shown that unique

spectral characteristics are observable, especially

from 650-800nm wavelengths. Figure 9 shows the

spectral characteristics of different azurite pigments.

They are grouped into natural and artificial mineral

pigments. It is noticeable that all the pigments have

similar peak positions around 430-480 nm, which

gives them bluish hue. However, starting from about

650 nm, the percentage reflectance of the artificial

pigments starts to increase drastically while the

natural pigments did not show significant change.

These spectral signatures may be used to identify the

nature of the pigments used in an artwork.

Figure 9: Spectral reflectance of natural and artificial

azurite mineral pigments. The alphanumeric symbol

enclosed in parentheses indicates whether the pigment is

natural or artificial while the other indicates relative

particle size: the higher the number, the smaller the

particle size.

3.3 Analysis of Cultural Heritage

The analysis of cultural heritage presents several

technical challenges. First, its delicate nature

requires non-destructive and non-invasive analysis.

The techniques presented in this paper satisfy the

requirement. However, there are some issues that

cannot be fully covered by a single technique. For

example, SRXRF method offers high accuracy and

precision but it can only give information about the

chemical composition of the pigments used in the

artwork. When it is used in the study of

discoloration of heated pigments, it cannot give

sufficient information. VL-NIS can give more

information in this case; the changes in color due to

heating can be tracked by the changes in spectral

reflectance. XAFS can also provide insight on the

discoloration by analyzing the chemical state of

copper absorption edge. However, neither SRXRF

nor XAFS can distinguish between the differences in

pigment size. It was observed that pigment particle

size affects the hue. Small pigments have lighter hue

than large pigments even though they have the same

chemical composition. This is illustrated by Figure

10 and confirmed by the SRXRF spectra given by

Figure 11.

IMAGAPP 2009 - International Conference on Imaging Theory and Applications

126

Figure 10: Pigment hues as a function of relative particle

size and heating temperature.

Figure 11: Corresponding SRXRF spectra.

Another concern that poses technical challenge

on the analysis of cultural heritage is the preparation

of sample. In advance analytical techniques such as

synchrotron radiation, the size of the sample is quite

small. In order to satisfy this, preparing a small

fraction of the sample is necessary. In this regard,

this can be considered destructive. In most cases,

this is unacceptable since a cultural heritage is

irreplaceable. On the other hand, VL-NIS can

address this concern. The size of the sample is

basically irrelevant. In VL-NIS, the major concern is

the accuracy. It is worth mentioning that the

limitation in sample size in X-ray techniques can be

addressed by using portable devices, such as

portable XRF. This is widely practice in analyzing

artworks. However, its accuracy is inferior

compared with large-scale X-ray devices. Therefore,

in the analysis of cultural heritage, the techniques

presented should be used in conjunction with one

another and not independently. This ensures that the

information extracted from the images can be

maximized.

4 CONCLUSIONS

Analysis of cultural heritage using analytical

imaging is a burgeoning field. This is attributed to

the numerous technical challenges it presents (i.e.

non-destructive and non-invasive). In this study,

synchrotron radiation and visible light-near infrared

techniques were used to analyze real cultural

heritage. It is considered that detailed analysis of

such historic objects (in this case, an ancient

Mongolian textile and old Korean painting) is of

prime importance. In addition, the mechanism of

color discoloration as a result of heating was

investigated. Synchrotron radiation analysis revealed

that it was due to the change in chemical bonding

state. VI-NI analysis also revealed some results. It

was shown that the change in chemical bonding state

results in increase or decrease in magnitude of the

spectral reflectance but does not affect the position

of the peaks. Having a clearer understanding of

pigment degradation can help in cultural heritage

preservation and restoration efforts. The results have

also shown that no single technique is capable of

providing all the necessary analytical information.

The available techniques should be used to

complement one another. It was also shown in this

study that analysis using visible light- near infrared

radiation could be an indispensable tool for

investigating cultural heritage. This is due to some

unique spectral features of materials that are only

observable in this range. Since VL-NIS does not

require complex instrumentation, robust and flexible

systems are achievable.

ACKNOWLEDGEMENTS

This work has been done as part of the project “An

Integrated System for Secure and Dynamic Display

of Cultural Heritage” sponsored by Japan Science

and Technology Agency, Regional Resources

Development Program. This collaborative project

was organized by Kyoto University Graduate School

of Engineering, S-tennine Kyoto (Ltd) and Kyushu

National Museum. The Authors would like to

express their thanks to Imazu Setsuo of Kyushu

National Museum and other staff of the museum

and, Oshima of S-tennine Kyoto and his group for

supporting this work. The authors are also grateful to

Prof. Atsuo Iida of Photon Factory (Tsukuba,

Japan). Thanks is also due to Mr. G. Enkhbat of

Center of National Heritage, Mongolia who supplied

the Mongolian textiles used as samples.

ANALYTICAL IMAGING OF CULTURAL HERITAGE BY SYNCHROTRON RADIATION AND VISIBLE LIGHT-

NEAR INFRARED SPECTROSCOPY

127

REFERENCES

Balas, C., Papadakis, V., Papadakis, N., Papadakis, A.,

Vazgiouraki, E., Themelis, G., 2003. A novel hyper-

spectral imaging apparatus for the non-destructive

analsyis of objects of artistic and histori values.

Journal of Cultural Heritage, 4, pp 330s-337s.

Fauble, W., Staub, S., Simon, R., Heissler, S., Pataki, A.,

Banik, G., 2007. Non-destructive analysis for the

investivation of decompodition phenomena of

historical manuscripts and prints. Spectrochimica Acta

Part B, 62, pp 669-676.

Ide-Ektessabi, A., 2007. Applications of Synchrotron

Radiation. Springer- Verlag Berlin Heidelgberg.

Margaritondo, G, 1988. Introduction to Synchrotron

Radiation. New York Oxford, Oxford Univeristy

Press.

Toque, J.A., Nishimura, R., Ide-Ektessabi, A., 2007,

Analysis of cultural heritage by synchrotron radiation

and visible light-near infrared spectroscopy. PF

Activity Report 2007, 4A, 25B/2006G110.

Toque, J.A., Sakatoku, Y., Komori, M., Murayama, Y.,

Ide-Ektessabi, A., 2008, Analytical imaging of cultural

heritage by UHRS and MBI, 1

AUN/SEED-Net

Regional Conference in Manufacturing Engineering,

Manila Philippines, November 24-25, 2008.

IMAGAPP 2009 - International Conference on Imaging Theory and Applications

128