Journal of Imaging Science and Technology® 52(1): 010201–010201-15, 2008.
© Society for Imaging Science and Technology 2008
Beyond Red–Green–Blue (RGB): Spectrum-Based Color
Imaging Technology1
Masahiro Yamaguchi
Imaging Science and Engineering Laboratory, Tokyo Institute of Technology, 4259-R2-56, Nagatsuta,
Midori-ku, Yokohama, 226-8503, Japan
E-mail: [email protected]
Hideaki Haneishi
Research Center for Frontier Medical Engineering, Chiba University, 1-33 Yayoi-cho, Inage-ku, Chiba,
263-8522, Japan
Nagaaki Ohyama
Imaging Science and Engineering Laboratory, Tokyo Institute of Technology, 4259-R2-56, Nagatsuta,
Midori-ku, Yokohama, 226-8503, Japan
Abstract. This article presents a comprehensive study on the
spectrum-based color reproduction system, called Natural Vision
(NV), which aims to break through the limitation of red–green–blue
(RGB) three-primary schemes. After a basic discussion on the motivation for color imaging technology beyond RGB, the method for
systematizing the multispectral and multiprimary color imaging technologies, including image capture, processing, storage, printing, and
display, is presented. Then experimental multispectral systems for
both still image and video are introduced, and the following features
of spectrum-based scheme are revealed: a) highly accurate color
reproduction is possible even under different illumination environment, b) an expanded color gamut can be reproduced by
multiprimary color displays, c) the influence of observer metamerism
can be reduced by the spectral color reproduction, and d) the quantitative spectral attributes of an object, useful for its analysis or recognition, can be captured and preserved. Finally, the effectiveness
of the system is also demonstrated through experiments in fields of
application, such as medicine, digital archives, color printing, electronic commerce, and computer graphics. © 2008 Society for Imaging Science and Technology.
关DOI: 10.2352/J.ImagingSci.Technol.共2008兲52:1共010201兲兴
INTRODUCTION
Digital images are widely used at present in digital broadcasting, digital still cameras (DSCs), and image delivery over
the broadband network and are being applied to extended
fields such as telemedicine, electronic commerce, and electronic museum. The natural color reproduction is one of the
key issues in these applications, in addition to highresolution or large-screen display technologies. However,
many of the conventional color imaging systems are designed for user preference, and thereby it is difficult to reproduce the original color of an object. Color management
technology has greatly progressed, especially in color print1
Presented in part at IS&T/SID’s Fourteenth Color Imaging Conference,
November 6–10, 2006, Scottsdale, AZ USA.
Received Jan. 19, 2007; published online Jan. 15, 2008.
1062-3701/2008/52共1兲/010201/15/$20.00.
ing, but there still remains a limitation in color reproducibility, employing red–green–blue (RGB) three primary color
systems.
An approach to break through this limitation is going
beyond RGB, i.e., using a spectrum-based system instead of
RGB. It has been reported that the use of multispectral imaging significantly improves the color accuracy.1–5 In the display industry, the multiprimary color approach becomes one
of the choices for expanding color gamut,6–12 but it is difficult to take full advantage of multiprimary color technology
because a wide gamut image source is not available for such
displays. A total system based on spectral information has
not yet been established. A few papers13–17 were devoted to
multispectral systems, including both input and output, for
hardcopy applications14,16,17 or still image display,13–15 but
no systematized implementation of the multispectral and
multiprimary color imaging technologies has been reported.
This article presents the concepts for systematization,
implementation, and applications of video and still image
systems with spectrum-based color image reproduction, including multispectral image capture, processing, compression, transmission, storage, printing, and display. The system
presented in this article enables exploitation of the advantages of multispectral and multiprimary technologies. In addition, conventional imaging devices based on RGB can be
also employed in the proposed spectrum-based system,
though the color reproduction accuracy is not very high.
With the integrated model system, the effectiveness of the
spectrum-based technology is also demonstrated in various
fields of application. The results shown in this article are
primarily obtained from the project Natural Vision (referred
to as NV hereafter), an industry-government-academic joint
project (1999–2006), aimed at the development of a visual
telecommunication system that enables high reality image
reproduction with natural color.18–21 Although some results
were previously published, this article comprehensively sum010201-1
Yamaguchi, Haneishi, and Ohyama: Beyond RGB: Spectrum-based color imaging technology
marizes the study on the spectrum-based system for color
image reproduction.
SPECTRUM-BASED COLOR REPRODUCTION
SYSTEM
Why is a Spectrum-Based System Needed?
Let us briefly review the limitations of conventional RGBbased systems in this section to make clear the advantage of
spectrum-based color reproduction.
(1) The RGB values obtained in conventional systems
often have different meanings, depending on the
device characteristics or color processing. For example, many conventional color imaging systems
are designed for user preference, and the RGB values do not represent objective color information.
Even if colorimetric color calibration is applied,
most color cameras do not satisfy the Luther condition; i.e., the spectral sensitivity is different from
that of human vision.22 Thus the RGB signal does
not have one-to-one correspondence to the
tristimulus values perceived by human vision. If the
spectral sensitivity is closer to that of human vision,
the color fidelity is improved, but noise behavior
may become worse. Spectral considerations are
thus essential in the color acquisition process.
(2) The RGB or any other three primary color signals
that comply with the color space, such as sRGB or
YCbCr, are defined under a white point such as CIE
D65 or D50 standard illuminant. When the illuminant of the observation environment is different from that of image capture, the color under the
different illuminant should reproduce the color as if
the object were placed at the site of the observer.
White balance adjustment in conventional color
imaging is performed in RGB space, sometimes introducing a color appearance model. But the colorimetric accuracy is not high, because the spectral
reflectance of the object and the spectral distribution of illuminant are required in principle in order
to calculate the color under the different illuminant. When images are displayed on softcopy
monitors, it is possible to reproduce the color under different illuminant based on spectrum-based
color conversion.18 However, there is an additional
issue: the standard color spaces are defined for a
certain standard illuminant and do not support the
color under an arbitrary illuminant. In the
hardcopy applications, moreover, spectral printing
is required to solve this problem of illuminant
metamerism.23
(3) In the color image display, the color gamut does
not cover all the existent colors, and some highsaturation colors cannot be reproduced. To enlarge
the color gamut, the saturation of primary colors
can be increased, but the gamut is still limited
within a triangle. The multiprimary color approach, i.e., using more than three primary colors,
010201-2
was proposed for a larger color gamut.6–8 The
gamut then becomes a polygon, where its vertices
correspond to the color coordinates of the primaries, or a polyhedron in three-dimensional color
space. Even if the display device allows the display
of a wider color gamut, conventional color signals
such as sRGB or ITU-R BT.709 do not support a
wider gamut color signal. Wide gamut color spaces
have recently become available,24 such as
AdobeRGB and xvYCC, but most of the color input
devices cannot capture high-saturation colors correctly for the reason explained previously.
(4) In highly accurate color reproduction, the effect of
observer metamerism cannot be ignored, which has
its origin in the difference of color matching functions among individuals. When the color displayed
on a monitor is compared with the real objects,
such as printed materials, the observed colors may
disagree with each other due to the observer
metamerism effect, even if the colorimetric accuracy is high. The multispectral and multiprimary
approach solves this problem by means of spectral
color reproduction.18,25
(5) In the image archive, database, or analysis, the utilization of color information is limited, since the
RGB signal depends on the devices, illuminants,
and preprocesses involved in the imaging systems.
For example, in the image retrieval using color information, the target object cannot be found if the
illumination condition is different. In contrast to
this, the original attribute of an object that generates color, i.e., spectral radiance, reflectance, or
transmittance information, is captured and preserved with multispectral imaging. The quantitative
color information is useful for the analysis or the
recognition of the object in the image database of
the digital archives26,27 of cultural heritage, artworks, and clinical cases in medicine. Moreover, the
exploration of invisible features becomes possible
from spectral images.
The spectrum-based color reproduction provides the solution to these problems, as shown in Figure 1, but it is necessary to integrate the system so as to take advantage of it in
image and video communication systems.
Background
It is known that spectral measurement provides accurate
colorimetry, and the spectral color acquisition is also utilized
in imaging applications. In addition to accurate imaging
colorimetry, multispectral imaging has been employed for
computational color constancy4,28 and color image
analysis.26,27,29,30
Although there have been numerous reports on theoretical and experimental aspects of multispectral imaging, few
works can be found on the system architecture for spectral
color imaging including input and output. A significant contribution to the system architecture was made by Keusen13
J. Imaging Sci. Technol. 52共1兲/Jan.-Feb. 2008
Yamaguchi, Haneishi, and Ohyama: Beyond RGB: Spectrum-based color imaging technology
Figure 1. The role of spectrum-based color reproduction in the video and still-image systems. 共a兲 Reproducing
the color as if the observer were at the remote site, 共b兲 reproducing the color as if object were placed at the
site of observer, 共c兲 expanding color gamut of the display 共the hexagon denotes the gamut of six-primary
projector兲, and 共d兲 enabling the storage and analysis of the image with quantitative spectral information.
and Hill.14 In the proposed architecture, the spectral stimulus, which is an output of the spectral scanner, is expanded
by a set of basis functions, and encoded as K-channel coefficients for the basis functions. The first three components of
the encoded data correspond to the tristimulus values under
a standard illuminant, and the remaining K-3 channels represent invisible spectral components. It was suggested to employ such spectral data for multispectral display and
printing.14
Rosen et al. revealed the concept of spectral color management including both input and output, and a spectral
profile connection space (PCS) was introduced instead of
the conventional PCS used in the color management
system.15,16 The architecture mainly deals with the spectral
printing, though an example of a spectral image visualization tool was reported.15 In addition, to reduce the dimensionality of the data for spectral printing, an interim connection space (ICS) was introduced, called LABPQR,17 where
the first three channels (LAB) corresponds the color space
under standard illuminant and the additional three channels
(PQR) corresponds to the “invisible” components. The concept is quite similar to the one of Keusen and Hill, but the
spectral PCS is a more flexible approach for the interchange
of spectral color information.
The spectrum-based color reproduction system described in the following section was also presented in this
context from the NV project.18,21 The architecture is similar
to the spectral PCS proposed by Rosen et al., and it is tested
in various application fields of multispectral imaging,
J. Imaging Sci. Technol. 52共1兲/Jan.-Feb. 2008
namely, the color reproduction in both still image and video
systems, multiprimary or spectral displays, printing, and
spectral image analysis. The architecture enables retention
and utilization of the information captured by multispectral
imaging in every class of applications. The feasibility of the
architecture has been confirmed through experiments in
various application fields, as described next. One of the specific features of the work in the context of the NV project is
the multispectral video system, in which real-time color conversion, transmission, and display is realized using the sixband camera and the six-primary display, as explained next.
Another feature is the demonstration of multiprimary displays, which makes it possible to reproduce high-reality images with natural color and wide color gamut.
How Does a Spectrum-Based System Work?
The concept of the spectrum-based color reproduction system is depicted in Figure 2. To obtain sufficient information,
a multispectral camera (MSC) is desired as an input device.
The profile of the input device, including spectral sensitivity,
tone curve, and dark current level of the camera, and the
spectral energy distribution of the illuminant are attached to
multispectral image (MSI) data. In image processing, transmission, and storage, the image is accompanied by the profile metadata, so that the spectral radiance, reflectance, or
transmittance can be retrieved. In the image display, the image of tristimulus values or spectral radiance is reproduced
on three-primary or multiprimary color displays with device
calibration. The system can also provide data to a color
010201-3
Yamaguchi, Haneishi, and Ohyama: Beyond RGB: Spectrum-based color imaging technology
Figure 2. The scheme of the spectrum-based color reproduction system.
printing system in the form of CIE XYZ tristimulus values
or the spectral reflectance under D50 standard illuminant.
The architecture of our spectrum-based color reproduction
system is similar to the ICC (International Color Consortium) color management system, but the PCS is based on a
physical model, i.e., spectrum-based PCS (SPCS) instead of
a color appearance model. This can be any CIEXYZ under
arbitrary illumination, spectral radiance, or spectral reflectance (or transmittance). The information required for the
forward or inverse transform to SPCS is held in the profile
data. The profile data format was defined as NV image data
format.18,21,31 The XML version NV format was also developed for easier handling.32 Table I shows the summary of the
data items used in the NV profile data by XML format.
Figure 3 shows the schematic presentation of the
workflow of the spectrum-based color reproduction system.
Once a MSI is captured by an input device, raw image data
and the characteristics of the input device are obtained. The
spectrum of the illuminant can be also attached to the data.
Applying the spectral estimation technique described in
“Spectral Estimation,” the matrix to estimate the spectral
reflectance, transmittance, or radiance is derived and attached to the profile data. Artificial spectral image is also
supported, where it is accompanied by the characteristics of
the virtual input device,33,34 or the spectral reconstruction
matrix. It is also possible to assign the matrix to estimate the
tristimulus values under a certain illuminant for convenience. To make use of the stored MSI data afterward, it is
often important to know the profile of the input device. For
this purpose, the profile data as shown in Table I are attached to the image data in the database or in the system for
interchanging images. In image display or printing, the device profile of the output device, as well as the illumination
spectrum of the assumed output environment, is presented
to the color conversion system. By this system, the color
reproduction under an arbitrary illumination environment
as shown in Fig. 1(b) is possible, while the standard illuminant (such as CIE D65 or D50) can be specified for
output to the conventional color display or printing devices.
010201-4
Accordingly, the system is capable of output of the image
data compatible with the standard color space.
In this system, the number of channels in the image
capture and the display are independent, and input and output devices with arbitrary numbers of channels can be employed. Color conversion can be performed without loss of
information contained in the MSI. Three-channel devices
can also be used in this system with proper device characterization, although there arise restrictions in the color reproduction capability or accuracy.
MULTISPECTRAL AND MULTIPRIMARY
TECHNOLOGY
Multispectral Image Capture
For high-fidelity color reproduction with the spectrumbased color reproduction system, it is crucial to increase the
number of bands in the image input device. There have been
various reports on devices for the acquisition of mulispectral
images,2–5,27–30,34 for example, a monochrome camera with a
rotating filter wheel,2,4,5 a grism (grating-prism),35 or a liquid crystal tunable filter.30 Capturing a multichannel image
with modulating illumination light is also a useful technique
to obtain the spectral reflectance of an object.26 The use of
dichroic mirrors to separate the spectral components provides highest signal-to-noise ratio, though the optical system
becomes relatively complex.36 In the experiment explained
next, the MSCs for still image and video developed in the
NV project are used.18,19,37
Figure 4(a) shows the 16-band MSC with a rotating
filter wheel developed in the NV project.18 The color reproducibility evaluated using a GretagMacbeth color checker is
shown in Figure 4(b). The accuracy by MSC is high and the
error is smaller than the discriminable level of human vision,
while visually apparent error is observed in three-band DSC.
Figure 5(a) shows a six-band HDTV camera for the
acquisition of motion picture.37 In the six-band camera, the
light is divided into two optical paths by a half mirror, and
incident on two conventional three–charge coupled device
(CCD) cameras after transmission through the special color
J. Imaging Sci. Technol. 52共1兲/Jan.-Feb. 2008
Yamaguchi, Haneishi, and Ohyama: Beyond RGB: Spectrum-based color imaging technology
Table I. Summary of the sample profile data in XML version of NV format. The mark … represents the omitted numerical data.
XML
Description
具Nvision xmlns“http:// … /NvXmlSchema.xsd”典
具NvisionImage典
具ImageInfo d2p1:ID=“1”典
具ImageType典RAWIMAGE具/ImageType典
具ImageWidth典1920具/ImageWidth典
具ImageHeight典-1080具/ImageHeight典
具ImageBands典6具/ImageBands典
具BitSizePerBand典16具/BitSizePerBand典
具DataType典UINT16具/DataType典
具/ImageInfo典
具/NvisionImage典
Basic information of image data
具NvisionInput d2p1:InputDate=“2006-12-31T00:00:00.0000000+09:00”典
Input profile tag
具DarkCurrentData d2p1:ID=“1”典
具DarkCurrentValue典5842.29 … 5113.90具/DarkCurrentValue典
具/DarkCurrentData典
N-Component bias digital counts
具ToneCurvesData d2p1:ID=“2”典
具CurveValues典0 0 … 65535具/CurveValues典
具/ToneCurvesData典
N-Component tone reproduction curves
具ExposureTimeData d2p1:ID=“3”典
具ExposureTimeValue典1.99 … 0.998具/ExposureTimeValue典
具/ExposureTimeData典
Coefficients of N-component
sensitivity level for the correction of
exposure time setting
具SpecSensiData d2p1:ID=“4”典
具SpecSensiValue典 0.000000 … 4.269989具/SpecSensiValue典
具/SpecSensiData典
Spectral Sensitivity of P-channel input
device
具NoiseData d2p1:ID=“5”典
具NoiseValue典0.699997 … 0.699997具/NoiseValue典
具/NoiseData典
N-channel noise levels of Input
device
具InputIllu d2p1:ID=“6” d2p1:ShortWaveLength=“380.000000”
d2p1:DataNumber=“401” d2p1:WaveInterval=“1.000000”典
具InputSpecData典 0.007034 … 0.167847具/InputSpecData典
具/InputIllu典
Spectral radiance of Q-Input
illuminants
具AutoCorrelationMatrix d2p1:ID=“7”典
具AutoCorrelationEigenValue典0.000000 … 0.256989 具/AutoCorrelationEigenValue典
具AutoCorrelationMatrixValue典0.000000 … 0.099686 具/AutoCorrelationMatrixValue典
具/AutoCorrelationMatrix典
具/NvisionInput典
Principal components of object’s
spectral reflectance
具NvisionConversion
d2p1:ConvDate=“2006-12-31T00:00:00.0000000+09:00”典
具XYZConvData d2p1:ID=“8”典
具XYZConvValues典0.208511 … -0.004578具/XYZConvValues典
具/XYZConvData典
Relative colorimetric values
estimation matrix
具SpecReflectData d2p1:ID=“9”典
具SpecReflectValues典0.642792 … 0.000366具/SpecReflectValues典
具/SpecReflectData典
Relative spectral reflectance
estimation matrix
具SpecStimuliData d2p1:ID=“10”典
具SpecStimuliValues典0.068008 … 0.000031具/SpecStimuliValues典
具/SpecStimuliData典
Relative spectral radiance
estimation matrix
具ColorMatchingFuncData d2p1:ID=“11”典
具ColorMatchingFuncValues典0.001358 … 0.000000具/ColorMatchingFuncValues典
具/ColorMatchingFuncData典
具RenderingIllu d2p1:ID=“12” d2p1:ShortWaveLength=“380.000000”
d2p1:DataNumber=“401” d2p1:WaveInterval=“1.000000”典
具RenderingSpecData典0.007019 … 0.167862具/RenderingSpecData典
具/RenderingIllu典
具/NvisionConversion典
filters inserted in each path. Six-channel uncompressed
HDTV video signal is captured and recorded on a magnetic
hard disk. As a result of evaluating the color reproducibility
of this six-band HDTV camera using the GretagMacbeth
color checker, the CIELAB color difference average and
*
*
maximum ⌬Eab
= 1.4 and 4.2, respectively, where ⌬Eab
= 4.1
and 8.2 in the case of the conventional HDTV camera. The
J. Imaging Sci. Technol. 52共1兲/Jan.-Feb. 2008
same Wiener estimation technique mentioned in the next
section was used in the evaluation of both devices.
Figure 5(b) shows the estimation results of high saturation colors. In this experiment, the interference filters of the
16-band MSC were illuminated by a light box (DNP,
Hi-Vision Color Viewer) and employed as test colors. The
colors were captured with the six-band camera and the
010201-5
Yamaguchi, Haneishi, and Ohyama: Beyond RGB: Spectrum-based color imaging technology
Figure 3. The schematic presentation of the workflow of multispectral images. The spectrum-based color
reproduction system supports the image management with the metadata 共profile data兲.
Figure 4. 共a兲 16-band MSC 共4M pixels/band兲. 共b兲 The color estimation accuracy of the 16-band MSC and
the 3-band DSC, under daylight 共DL兲, CIE A and F2 illuminants, where DL was used in the image capturing.
Gray and white bars denote the average and maximum CIELAB color differences, respectively.
tristimulus values were calculated by the spectrum-based
method explained next. The results are compared with the
colors measured by a spectroradiometer in Fig. 5(b). The
simulated results with a three-band camera are also shown
in the Fig. 5(b). It can be seen that the colors estimated from
the three-band camera clusters around RGB pure primaries,
that is, the vertices of the camera analysis gamut. The six010201-6
band camera offers considerably better color tones in these
high-saturation colors in comparison to the three-band
camera. The color differences are relatively large for green
objects, but they are out-of-gamut of the output devices,
even in the case of wide-gamut displays. The results clearly
show the advantage of multispectral imaging, especially in
the case of high-saturation colors. The camera was employed
J. Imaging Sci. Technol. 52共1兲/Jan.-Feb. 2008
Yamaguchi, Haneishi, and Ohyama: Beyond RGB: Spectrum-based color imaging technology
more easily handled devices will become available for the
measurement of the illumination spectrum.
Figure 5. 共a兲 6-band HDTV camera. 共b兲 Color estimation results of the
colors of interference filters by the six-band camera 共쎲兲 in CIE xy chromaticity diagram. The ideal color coordinates obtained by spectral measurements 共⫻兲 and the simulated results obtained by the three-band camera
共䉭兲 are also shown. Black and gray arrows denote the color shifts from
the original colors for six-band and three-band cases, respectively.
Dashed lines represent the locus of the colors obtained from the six-band
camera.
for the experiment of real-time video reproduction described next.
In image capture, the spectral distribution of the illumination is also measured to estimate the spectral reflectance
or the transmittance of the object. For example, the reflected
light from a standard white plate is measured by a
spectroradiometer or the MSC, or the ambient illumination
is directly measured by a compact fiber-optic spectrophotometer. For correction of the spatial nonuniformity of illumination, a white plate (in the case of reflective object) or a
transparent glass plate (in the case of transmissive object) is
captured by the MSC. For practical use, it is expected that
J. Imaging Sci. Technol. 52共1兲/Jan.-Feb. 2008
Spectrum Estimation
In the spectrum-based color reproduction system, the spectral radiance, reflectance, or transmittance is estimated from
the MSI data. An arbitrary estimation method can be used
in this system, as the original MSI data can be preserved
with the spectral sensitivity of MSC. Among various techniques for spectral estimation, the method based on Wiener
estimation is practical and accurate. If the covariance function of spectral reflectance of the target object is available, it
considerably improves the accuracy. However, it is not necessarily possible to obtain the covariance function, and we
can alternatively use an approximation in which the covariance is modeled as a first-order Markov process.1 This
model works fairly well for most natural objects, because the
spectral reflectance of natural objects is for the most part
smooth, except in special cases.
Even though spectrum estimation may be used in the
color conversion where the spectrum space requires high
dimensionality, the implementation of the spectrum-based
color conversion is not so difficult in many cases, insofar as
the spectral estimation may be be formulated by a linear
model.38 For example, to display the color under an illuminant different from the image capturing environment,
the image spectral reflectance is estimated from the multispectral image and the illuminant spectrum of the image
capturing environment. Then the tristimulus values are calculated by multiplying the illuminant spectrum of image
display environment and the color matching function. Then
the device RGB signal is obtained by multiplying the matrix
for color conversion. After the nonlinearity of the image
capturing device is corrected, the spectrum-based color conversion from the multispectral image to the device RGB signal can be realized with the matrix operation of N ⫻ 3, and
following tone curve correction which can be implemented
by one-dimensional look-up tables. Therefore, computation
in the spectral space is needed only when changing the color
conversion matrix.
In color reproduction for video images, the gain parameter noticeably impacts the color accuracy and the image
quality. The gain parameter should be set such that the sixband signals from a reference white object become constant,
in order to set the signal levels within the dynamic range of
every band. Then the gain parameters are required in the
color estimation step to correct the magnitudes of the output signals, as well as to set the noise variance in the derivation of a Wiener estimation matrix, since the noise levels
of the camera output depend on the gain setting.
Multiprimary Display System
There were few previous works on multiprimary color displays for larger color gamut; NHK (Japan Broadcasting
Corp.) demonstrated a four-primary projection display to
evaluate the wider color space in 1994.6 The present authors
also reported a seven-primary display using a holographic
optical element.7 However, system development and evalua010201-7
Yamaguchi, Haneishi, and Ohyama: Beyond RGB: Spectrum-based color imaging technology
Figure 6. The six-primary projection display using two modified projectors. The spectral intensities of each
projector are also shown.
Figure 7. Color gamut of six-primary DLP display, conventional three-primary DLP display, and natural object,
in a*-b* planes of L* = 10, 50, and 90.
tion of the muliprimary display had been originally started
in the context of the NV project. In the multiprimary projection display developed in the NV project, the images from
two projectors [liquid crystal display (LCD) or digital light
processing (DLP)], each of which is adapted to produce a
different set of three primary colors, are overlaid on the
screen, as shown in Figure 6.8,9 As a flat-panel type
multiprimary display, a four-primary flat-panel LCD with
light emitting diode (LED) backlight was implemented by
means of time sequential display.10
Figure 7 shows the color gamut of a six-primary DLP
display.9 As for the gamut of natural objects, Pointer gamut39
and standard object color spectra (SOCS) database (excluding fluorescent objects)40 are combined 共Pointer+ SOCS兲,
because some objects in SOCS are out of the Pointer gamut.
Six-primary DLP display almost covers the gamut of natural
objects and the gamut volume in CIELAB space is about 1.8
times larger than that of the normal RGB DLP projector.
The gamut of the six-primary display is enlarged in the dark
red, cyan, purple and bright orange regions, as compared
with the conventional RGB display.
010201-8
Spectral Color Display
There exists variability in color matching functions of human observers originating from macular pigments, lens absorption, and cone sensitivity. Due to the individual difference of color matching functions, a color difference may
appear to a certain observer when two color stimuli are
shown, even if perceived as the same color by another observer, or the CIE standard observer. This phenomenon is
called observer metamerism. It causes the color mismatch
between different media, such as color printed materials and
displays, even though the colorimetric match is achieved for
CIE 1931 or 1964 color matching functions. It has been
shown that the influence of observer metamerism due to the
variation of color matching functions in color reproduction
is not negligible.41,42
Based on the multispectral and multiprimary technology, spectral color display becomes possible, and the color
mismatch between the display and the real object can disappear for different observers. The advantage of spectral color
display was experimentally demonstrated using the six-
J. Imaging Sci. Technol. 52共1兲/Jan.-Feb. 2008
Yamaguchi, Haneishi, and Ohyama: Beyond RGB: Spectrum-based color imaging technology
Figure 8. The system architecture of multispectral video transmission. Case 1: Multispectral input signal is
transmitted, and the color under the arbitrary illuminant can be computed at the receiver site. This scheme is
suitable for image archive and distribution purposes. Case 2: The spectrum of observation illuminant is passed
to the sender in advance and the color under the observation illuminant is transmitted. Suitable for one-to-one
communication, or natural color reproduction with fixed observation illuminant. Case 3: The spectrum of
observation illuminant and the display profile are passed to the sender in advance, and the display signal is
transmitted. This seems suitable for client-server image transmission.
primary display25 with the signal processing method briefly
explained in the next subsection.
Multiprimary Color Conversion
The multiprimary color signal is generated from the image
of tristimulus values or multispectral data, called
multiprimary color conversion, similar to the color decomposition for multiprimary color printers. For colorimetric
color reproduction, three-dimensional tristimulus values are
transformed to M-dimensional multiprimary color values. It
involves a degree of freedom; plural combinations of
multiprimary color values can reproduce a certain color.
The following methods have been developed as the
multiprimary color conversion for colorimetric color reproduction:
(1) The matrix switching method43 in which the polyhedral color gamut spanned by multiprimary colors
is divided into pyramids, and selecting a 3 ⫻ 3 matrix depending on the pyramid, we perform a linear
color conversion in each pyramid.
(2) Linear interpolation in equiluminance plane
method44 where the multiprimary signal values are
interpolated within the equiluminance plane in the
solid color.
(3) Metameric black method45 whereby multiprimary
signal values are decomposed into visible and invisible (metameric black) components. The visible
components are uniquely solved, where the
metameric black components are determined such
that the multiprimary signal values changes
smoothly if the change of tristimulus values is
smooth.
J. Imaging Sci. Technol. 52共1兲/Jan.-Feb. 2008
(4) The spherical average method46 in which the conversion is based on the analytical solution such that
the multiprimary signal values are continuous for
the smooth change of tristimulus values.
The computation speed, ease of hardware implementation, and image quality depend on the conversion method.
When an image with smooth color tonal change is reproduced on a multiprimary color display, a contourlike pattern
sometimes appears. The observer dependence of color
matching functions and the device characterization error are
the sources of the artifact, and proper selection of the
multiprimary color conversion process suppresses such
artifacts.47 Color tone reproduction is one of the important
issues for image quality in multiprimary displays.
For the spectral color reproduction explained in the
previous subsection, it becomes N to M conversion, where N
is the number of channels of MSI. In the proposed N to M
conversion method called the “spectral approximation
method,” the spectral error is minimized under the constraint that the colorimetric match is attained for the standard observer. The effectiveness of this approach was confirmed experimentally.25,48
Multispectral Image Transmission, Compression
The system architecture of image transmission and signal
processing in the spectrum-based color reproduction system
is shown in Figure 8. The signal from MSC (multispectral
input signal) is converted to a signal for transmission, and
converted again to the display signal for multiprimary or
RGB display. The same scheme can be used in the storage of
image data. There Three cases to implement the image
010201-9
Yamaguchi, Haneishi, and Ohyama: Beyond RGB: Spectrum-based color imaging technology
Table II. Transmission of video signal in natural vision
Case
Transmissionsignal
Description
Examples of application
1
Multispectral input signal
The color under the arbitrary illuminant can be computed at the receiver site
Image archive
Image distribution
2
Colorimetric signal
The spectrum of observation illuminant sent to the sender in advance and the color
under the observation illuminant is transmitted.
One-to-one communication
Wide-gamut color reproduction
with fixed observation illuminant
3
Multiprimary display signal
The spectrum of observation illuminant and the display profile sent to the sender in
advance, and the display signal transmitted.
Client-server image transmission
Figure 9. The experimental prototype system for multispectral and multiprimary video reproduction.
transmission system can be considered, and the features of
these implementations are given in Table II.
The compression and encoding are also issues for MSI
transmission. As an MSI compression technique considering
the colorimetric accuracy, a modified Karhunen-Lueve (KL)
transform called weighted KL transform was proposed and
combined with the JPEG2000 scheme.49 For video compression, three methods that support multichannel images,
MPEG4 studio profile, H.264/AVC, and Motion JPEG2000,
were tested. The test included a coding method in which a
MSI signal was converted into visible and invisible components, with the subsampling of chrominance signals. As a
result, it was shown that intraplane codecs and the fullresolution formats without subsampling give better quality
at higher bit rate, while a subsampled format surpasses other
methods at lower bit rate.50 JPEG2000 is one of the suitable
formats for high-quality encoding of both video and still
MSIs. The profile data for NV format described previously
can be easily implemented in JPEG2000 as an extended ICC
profile or metadata using XML.
Real-Time Video Reproduction and Transmission System
A prototype of the multispectral video transmission shown
in Figure 9 was also developed.51 The six-channel video signal from six-band high definition television (HDTV) camera
is converted in a six-to-three color converter to the colorimetric tristimulus values, by using a spectrum-based color
reproduction technique. In the spectrum-based color reproduction, after correction of the tone reproduction curve of
CCD, the spectral reflectance is calculated for every pixel by
010201-10
Wiener estimation with the spectral sensitivity of the camera
and illuminant spectrum of the image-capturing environment measured in advance. The colorimetric tristimulus values are then generated with using the illuminant spectrum
of the observation environment. All of these processes can
be done by 6 ⫻ 3 matrix multiplication in real-time.
For the image display, both three-primary and
multiprimary displays can be employed. The color display is
characterized in advance and the parameters are set into the
real-time color converter. When using the six-primary color
display, a three-to-six color converter consisting of six 3D
LUTs is used to generate a six-primary color video signal.
Figure 10 is the photograph of the demonstration of sixband real-time reproduction, along with comparison with
the real objects and the three-band system. The advantage of
the spectrum-based system can be easily observed from the
demonstration in Figure 10.
In real-time reproduction, direct conversion from three
to six channels is difficult because of the hardware complexity. Thus colorimetric color reproduction is implemented
with the six-to-three color converter. In addition, it is possible to apply a limited version of the spectral approximation
technique; the spectral radiance of the object is approximated from the colorimetric tristimulus signal using the
spectral estimation technique with the illuminant spectrum
of observation environment. The object-dependent basis
functions can be employed to reconstruct the spectra from
the tristimulus values if available. In this case, the spectral
difference between the real object and the displayed image
J. Imaging Sci. Technol. 52共1兲/Jan.-Feb. 2008
Yamaguchi, Haneishi, and Ohyama: Beyond RGB: Spectrum-based color imaging technology
Figure 10. Experimental live reproduction using a six-band HDTV camera, six-primary display, and real-time color conversion, comparing with
conventional HDTV.
becomes smaller if the spectral shape of the reflectance is
smooth, which is assumed in the spectral estimation.
Through the visual observation in an informal manner, the
effectiveness of the spectral approximation is found by direct
comparison of the object and the display, for example, in the
reproduction of skin colors.
The captured image is recorded on the hard disk drive
(HDD) video recorder without any compression. The data
volume is about 3.5 Gbps. In an experiment on real-time
reproduction at a remote site, the parallel MPEG-2 encoder/
decoder for transmitting video signal through the TCP/IP
network was used. When the network bandwidth was
80 Mbps, the CIELAB color difference due to the codec and
*
the network was smaller than unity 共⌬Eab
兲.
Multispectral Image Editing, Analysis, and Database
A multispectral video editor is required to create video contents. In the multispectral video editor, it is expected to
handle image data of an arbitrary number of spectral bands.
It is possible to edit movies from MSI sequences of the various numbers of bands using a sort of script, but it becomes
rather burdensome to handle different format images. A
method was proposed to simplify the multispectral video
editor in which the image data of various numbers of bands
are converted to a single format with the fixed number of
bands, where a virtual multispectral camera is assumed to
define the fixed format.33 A MSI can be considered as a
value-added color image because every pixel has quantitative
spectral data. Thus it is possible to make use of MSI data
with the image database and analysis. The techniques for
spectrum-based image processing, analysis, and retrieval explained in the next section were developed for specific applications, but they can be exploited in other applications as
well.
APPLICATIONS
Medical Application
The use of digital color images is extremely valuable for
teleconference, teleconsultation, education, training, image
analysis, and reference database in pathology, endoscopy,
and dermatology. In these applications, it may also be reJ. Imaging Sci. Technol. 52共1兲/Jan.-Feb. 2008
quired to reproduce the complexion of a patient through the
visual communication system for telediagnosis and home
care. If the reproduced color is not accurate, it may cause
misdiagnosis, thus, the reliability of reproduced color is critical.
In the pathology application, the spectrum-based system enables not only the accurate color reproduction of
stained tissue samples, but also the correction of the color
variation caused by the staining procedure. By use of the
spectral information in the image, the amount of dye in each
pixel can be estimated based on the Lambert–Beer Law, and
an image can be reconstructed by digital adjustment of the
density and balance of the staining.52 In addition, the spectral information is valuable for the feature extraction or image segmentation.30 To show the results of the image analyses in a form more familiar to medical doctors, a method
called ⬙digital stain⬙ was proposed.53 A colorization matrix
was applied to the results of pixel classification, and an image equivalent to a physically stained specimen was generated through computer processing from the MSI data. It
becomes easier to evaluate the result of image analysis using
this technique in the pathology field.
In dermatology, it has been shown that MSI is useful for
the diagnosis of melanoma.29 The experiments in the NV
project dealt mainly with inflammatory skin lesions, in
which a faint color difference is important for diagnosis. The
MSI systems of both still image and video were tested, and
the color reproducibility of the multispectral system was
shown to be sufficiently high from the visual evaluation by
dermatologists.54,55 It was also shown that spectral information can be utilized for image analysis that supports diagnosis in both pathology and dermatology, for example, the
grading of disorders, or the quantitative evaluation of
treatments.54
Digital Archive and Electronic Museum
The color of archived images depends on the imaging device
and illumination condition, if we use conventional imaging
devices. Many reports are found on the multispectral imaging for digital archives.26,27,56 In the NV project, the digital
archives created both by multispectral still imaging and
video are investigated.
For the still image digital archive, the woodprint by
Shiko Munakata, a famous Japanese printmaker, was captured by the 16-band MSC, displayed on a screen, and transferred to the press in collaboration with Aomori Digital Archives Association. Moreover, with Biblioteca Nacional de
Antropologia y Historia (BNAH) and Instituto Nacional de
Astrofisica, Optica y Electronica (INAOE), Mexico, 16-band
MSIs of “codices,” pictorial documents of the Aztecan era,
were captured. The result of color reproduction displayed
using a 16-band multispectral image was found to be satisfactory for the art management staffs.57 In the capturing of
historical works, it is often necessary to minimize light irradiation, so the flash lamp, whose spectral energy distribution
had been measured in advance, was synchronized with 16
exposures in order to minimize exposure. For large size ob010201-11
Yamaguchi, Haneishi, and Ohyama: Beyond RGB: Spectrum-based color imaging technology
jects, a high-resolution image is similarly generated by tiling
smaller images.
The image data can also be employed for high-fidelity
color prints as explained in the next subsection. Furthermore, the spectral information in the image will be useful
for the analysis of pigments or dyes, or the selection of materials for restoration.
High-fidelity natural color reproduction will be valuable
for educational video and multimedia contents, promotions
of a regional nature, and virtual travel simulation. Multispectral movies were created to demonstrate the feasibility of
the multispectral video system, using the video editing technique explained previously. In the introduction of the technology developed in the NV project as an example of scientific video, the effectiveness of multispectral and
multiprimary technologies were demonstrated using a widegamut multiprimary display. The content is composed of
both the computer graphic (CG) images and the video captured by the six-band camera. An introduction of Okinawan
regional nature and culture related to color included the
aerial view video captured by the six-band video camera
from a helicopter. This material successfully demonstrated
the feasibility of the basic technology of the video acquisition and production, though improvement in usability is
strongly recommended.
In addition, the natural color reproduction of the aurora was tried experimentally in cooperation with the University of Alaska at Fairbanks. Since the luminance of the
aurora is very low, multispectral image capture is difficult.
Instead, high-sensitivity RGB cameras (NAC Image Technology Inc., with Texas Instruments Impactron CCD sensors),
are employed with spectral characterization. The spectral
characteristics of the aurora are caused by the oxygen and
nitrogen ions, which can be measured by spectroradiometer,
and these data are used in the spectral estimation from
three-band images. The images, 30 fps video images from
three cameras, 640⫻ 480 pixels for each camera, are tiled to
obtain a larger field of view. In the display, neutral density
filters are attached to the multiprimary projectors to reproduce the very low luminance of real aurora. By visual evaluation, the reproduced colors, including faint green, dark red,
and light pink, appear faithful to reality.
Printing Application
Multispectral imaging greatly improves the fidelity of color
prints. As a test for the printing of catalogs for product
promotion, assuming a normal photographic situation for
the commercial products, a 16-band MSC and professional
RGB DSCs were compared in joint work with DNP Media
Create Co., Ltd.58 In the current catalog printing process, the
color adjustment by in-house color proof is essential because
the color reproduction capability of conventional DSCs is
not enough. If the in-house color proofing and adjustment
can be omitted, the simplification of the printing workflow
becomes possible along with print quality improvement. In
the experiment, the object and camera setting were prepared
assuming normal conditions for photography of commercial
products by a professional photographer. Nine different ar010201-12
rangements were tried. A 16-band MSC and RGB DSCs
(Sinarback 23 by Sinar, and D1X by Nikon) were used for
comparison purposes. In the multispectral image capture,
three flash lamps were synchronized with shutters for 16
exposure times. At first, the images obtained from the RGB
DSCs were printed by an ink jet printer, which is used for
normal color proofing. A professional print director puts
instructions for color corrections on the printed samples.
Color correction instructions were given at four to five
points on average for every picture.
To process the image captured by 16-band MSC, the
spectral distribution of the flash lamp was measured in advance, and then the spectrum-based color reproduction applied to obtain the image in CIEXYZ color space under standard D50 illuminant. Then it was converted to 16-bit
CIELAB digital color, the ⬙unsharp mask⬙ filter was applied
to adjust the sharpness to the DSC images, and RGB data
were generated by Apple ColorSync 3.0 using an ICC profile
prepared for this purpose. The printed results were evaluated
again by the same print director, and no instruction for color
correction was given for any of the resultant samples. The
print director also commented that the printed results from
16-band MSI was even better than the proof for presentation
produced from RGB DSC image after color correction.
The development of spectral printing techniques, which
is expected to reduce the illumination metamerism, has been
reported. Although it is not directly dealt with in the NV
project, the spectrum-based color reproduction scheme explained in “Spectrum-Based Color Reproduction System”
supports the use of MSIS for spectral printing. For example,
spectral color management was proposed using the interim
connection space called LabPQR.17 It can be employed with
the system described in the section, Spectrum-Based Color
Reproduction System, above, by incorporating the conversion from the SPCS to LABPQR.
Furthermore, spectral color display can be useful for
color proofing applications using softcopy monitors, since
the color matching between display and print is considerably
improved thanks to spectral approximation.
Electronic Commerce Application
In electronic commerce (EC), such as online shopping using
color images, color differences may lead to the return of
purchased items. Furthermore, the color matching of
samples between designers, factories, buyers, and sales is
needed, and quantitative color information is quite beneficial. A high-fidelity natural color reproduction system
should contribute to the further evolution of EC for vehicles,
apparel, cosmetics, toys, and interior furnishings. A webbased server-client prototype of the electronic catalog system
shown in Figure 11 was developed in the NV project. Besides
color reproduction under an arbitrary observation illuminant, the features offered by the prototype system include automatic selection of the color patch index corresponding to the specified image, and the retrieval of a fabric
image based on the color histogram or spectral reflectance.
The display of a wider color gamut is required in textile
applications. From the experiment using 464 colors from
J. Imaging Sci. Technol. 52共1兲/Jan.-Feb. 2008
Yamaguchi, Haneishi, and Ohyama: Beyond RGB: Spectrum-based color imaging technology
Figure 11. An electronic commerce prototype system. Image data can be retrieved from the database and the
color under the observation illuminant can be reproduced through a web-based user interface. The illumination
environment and display profile are sent to the server, and the image for display is generated in the server and
sent back to the client.
SCOTDIC(R) color book (Kensaikan International Ltd.) for
cotton, it was confirmed that 100% of the colors are covered
by the six-primary display, while 6% are out of the sRGB
color gamut. The advantage of the six-primary displays becomes clearer in consideration of the character of ambient
light. According to the evaluation by color experts in the
apparel industry, the color difference between the real fabric
and the reproduced image is almost negligible, though the
impression of surface texture is somewhat different between
the display screen and the real fabrics. The use of motion
pictures improves such impression, better reproducing the
angular dependency of reflection in textiles, as well as in
metallic or pearl automotive finishes.
Wide Gamut CG System
The expanded color gamut of the multiprimary display provides a new tool or new colors for graphics expression.59
Through the experiences of graphic expression with the
high-chroma colors enabled by six-primary display, the significance of wide-gamut graphics was observed. This significance includes: (a) rendering of the reality of actual objects,
such as flowers, butterflies, marine blue water verdant green
leaves, dresses, and accessories, in vivid or splendid colors,
(b) the enhanced reality based on “memory color,” which
often shifts to higher saturation with time, (c) strong impact
or fantastic impressions resulting from color combinations
scarcely encountered in the real world, and (d) rich color
tones thanks to the expansion of the color range, including
depth impression by the subtle color change in these dark
colors, or the expression of gloss, metallicity, or emissive
colors.
Digital Prototyping—Spectral BRDF Measurement and
Rendering
Digital prototyping is an important technology for the expedition and efficiency of product development, and CG technologies enable the realistic rendering of virtual products,
using the bidirectional reflectance distribution function
(BRDF) or Bidirectional Texture Function (BTF), but the
color often disagrees with that of the real products. The
multispectral BRDF/BTF measurement system and multispectral rendering technique were developed for high-fidelity
color digital prototyping. The system is based on a device for
the BRDF measurement (OGM-3, Digital Fashion Ltd.), and
Xenon light sources to which filter wheels are attached to
capture 16-band MSI. The spectral BRDF rendering is realized using the multispectral image; the image of each band is
J. Imaging Sci. Technol. 52共1兲/Jan.-Feb. 2008
independently calculated. The result is reproduced on the
six-primary display and compared with the real object to
confirm high color reproducibility.34
Conclusions
The concept, technology, and applications of a spectrumbased color reproduction system have been introduced in
this article. Spectrum-based color reproduction enables not
only high-fidelity color reproduction but also the application
of image analysis based on quantitative spectral information.
Moreover, multispectral information will also be of great
utility in image editing for preferred color and other various
image processing applications such as object extraction or
image synthesis, though those were not the main topics of
the NV project, which basically targets natural color reproduction. Wide gamut display is one of the most recent topics
in the display industry, and presentations of multiprimary
displays can also be found from other groups. But conventional color camera devices and image color spaces such as
sRGB do not support wider gamut image data at present.
Going beyond RGB, significant benefits emerge in advanced imaging applications. Multiprimary printing is already available in commerce, but to make better use of devices that support innovative color reproducibility, a
platform for the spectrum-based system is expected, i.e.,
multispectral and wide-gamut video content creation, management, distribution, and utilization.
Acknowledgments
The authors acknowledge all the former research team members: Hiroyuki Fukuda, Norihito Fujikawa, Junko Kishimoto,
Hiroshi Kanazawa, Masaru Tsuchida, Ryo Iwama, Kenro
Ohsawa, Toshio Uchiyama, Hideto Motomura, Satoshi
Nambu, Tatsuki Inuzuka, Yuri Murakami, the research fellows, joint researchers of NV project, and the participating
companies: NTT, NTT Data, NTT Communications,
Olympus, NHK, Hitachi, JVC, Dainippon Printing, Toppan
Printing, Panasonic, Keisoku Giken, and Digital Fashion.
Special thanks are given to the following cooperating institutes and companies: Kagawa University Hospital, Yokohama
City University Hospital, University of Pittsburgh Medical
Center, BNAH, INAOE, the University of Alaska at
Fairbanks, Aomori Digital Archives Association, and DNP
Media Create Co., Ltd. This work is supported by the
National Institute of Information and Communications
010201-13
Yamaguchi, Haneishi, and Ohyama: Beyond RGB: Spectrum-based color imaging technology
Technology, the Ministry of Internal Affairs and Communication, and the Natural Vision Promotion Council of Japan.
REFERENCES
1
W. K. Pratt and C. E. Mancill, “Spectral estimation techniques for the
spectral calibration of a color image scanner,” Appl. Opt. 15, 73–75
(1976).
2
B. Hill and F. W. Vorhagen, “Multispectral image pick-up system,” US
Patent 5,319,472 (1994).
3
P. D. Burns and R. S. Berns, “Analysis of multispectral image capture,”
Proc. IS&T/SID 4th Color Imaging Conference (IS&T, Springfield, VA,
1996) pp. 19–22.
4
S. Tominaga, “Multichannel vision system for estimating surface and
illumination functions,” J. Opt. Soc. Am. A 13, 2163–2173 (1996).
5
M. Yamaguchi, R. Iwama, Y. Ohya, T. Obi, N. Ohyama and Y. Komiya,
and T. Wada, “Natural color reproduction in the television system for
telemedicine,” Proc. SPIE 3031, 482–489 (1997).
6
R. Yajima, M. Kanazawa, and S. Sakaida, “Wide-color-gamut system,”
ITE ‘94 Annual Convention (Institute of Television Engineers of Japan,
Saga, Japan, 1994) pp. 355–356 (in Japanese).
7
M. Yamaguchi, T. Ajito, and N. Ohyama, “Multiprimary color display
using holographic optical element,” Proc. SPIE 3293, 70–77 (1998).
8
T. Ajito, T. Obi, M. Yamaguchi, and N. Ohyama, “Expanded color gamut
reproduced by six-primary projection display,” Proc. SPIE 3954,
130–137 (2000).
9
H. Motomura, N. Ohyama, M. Yamaguchi, H. Haneishi, K. Kanamori, S.
Sannohe, “Development of six-primary HDTV display system,” Proc.
Int. Display Research Conference (SID, San Jose, CA, 2002) pp. 563–566.
10
S. Komura, I. Hiyama, and N. Ohyama, “Four-primary-color LCD for
natural vision,” Inf. Disp. 8(3), 18–21 (2003).
11
S. Roth, I. Ben–David, M. Ben–Chorin, D. Eliav, and O. Bben–David,
“Wide gamut, high-brightness multiple primaries single panel projection
displays,” SID Int. Symp. Digest Tech. Papers (SID, San Jose, CA, 2003)
pp. 118–121.
12
H. Sugiura, H. Kaneko, S. Kagawa, J. Someya and H. Tanizoe, “Sixprimary-color LCD monitor using six-color LEDs with an accurate
calibration system,” Proc. SPIE 6058, 60580H1–60580H8 (2005).
13
T. Keusen and W. Praefcke, “Multispectral color system with an
encoding format compatible with the conventional tristimulus model,”
Proc. IS&T/SID 3rd Color Imaging Conference (IS&T, Springfield, VA,
1995) pp. 112–114.
14
B. Hill, “Multispectral color technology: a way towards high definition
color image scanning and encoding,” Proc. SPIE 3409, 2–13 (1998).
15
M. Rosen, M. D. Fairchild, G. M. Johnson, and D. R. Wyble, “Color
management within a spectral visualization tool,” Proc. IS&T/SID 8th
Color Imaging Conference (IS&T, Springfield, VA, 2000) pp. 75–80.
16
M. Rosen, F. Imai, X. Jiang, and N. Ohta, “Spectral reproduction from
scene to hardcopy II: Image processing,” Proc. SPIE 4300, 33–41 (2001).
17
M. Derhak and M. Rosen, “Spectral colorimetry using LabPQR—an
interim connection space,” Proc. IS&T/SID 12th Color Imaging
Conference (IS&T, Springfield, VA, 2004) pp. 246–250.
18
M. Yamaguchi, T. Teraji, K. Ohsawa, T. Uchiyama, H. Motomura, Y.
Murakami, and N. Ohyama, “Color image reproduction based on the
multispectral and multiprimary imaging: Experimental evaluation,”
Proc. SPIE 4663, 15–26 (2002).
19
M. Yamaguchi, H. Haneishi, H. Fukuda, J. Kishimoto, H. Kanazawa, M.
Tsuchida, R. Iwama, and N. Ohyama, “High-fidelity video and stillimage communication based on spectral information: Natural vision
system and its applications,” Proc. SPIE 6062, 60620G1–60620G12
(2006).
20
M. Yamaguchi, “Beyond R.G.B.: Spectrum-based color imaging
technology,” Proc. IS&T’s NIP 22 (IS&T, Springfield, VA, 2006), pp.
326–330.
21
K. Ohsawa, H. Fukuda, Y. Komiya, M. Yamaguchi, and N. Ohyama,
“Spectrum-based color reproduction system for motion picture,” Proc.
SPIE 5008, 229–236 (2003).
22
P.-C. Hung, “Colorimetric calibration for scanners and media,” Proc.
SPIE 1448, 164–174 (1991).
23
L. A. Taplin and R. S. Berns, “Spectral color reproduction based on a
six-color inkjet output system,” Proc. IS&T/SID 9th Color Imaging
Conference (IS&T, Springfield, VA, 2001) pp. 209–213.
24
IEC61966-2-4, Multimedia
systems
and
equipment—Colour
measurement and management—Part 2–4: Colour management—
Extended-gamut YCC colour space for video applications—xvYCC
(2006).
010201-14
25
Y. Murakami, J. Ishii, T. Obi, M. Yamaguchi, and N. Ohyama, “Color
conversion method for multi-primary display for spectral color
reproduction,” J. Electron. Imaging 13, 701–708 (2004).
26
K. Martinez, J. Cupitt, D. Saunders, and R. Pillay, “Ten years of art
imaging research,” Proc. IEEE 90, 28–41 (2002).
27
H. Maitre, F. Schmitt, J. Crettez, Y. Wu, and J. Y. Hardeberg,
“Spectrophotometric image analysis of fine art paintings,” Proc. IS&T/
SID 4th Color Imaging Conference (IS&T, Springfield, VA, 1996) pp.
50–53.
28
L. T. Maloney and B. A. Wandell, “Color constancy: a method for
recovering surface spectral reflectance,” J. Opt. Soc. Am. A 3, 29–33
(1986).
29
B. Farina, C. Bartoli, A. Bono, A. Colombo, M. Lualdi, G. Tragni, R.
Marchesini, B. Farina, C. Bartoli, A. Bono, A. Colombo, M. Lualdi, and
G. Tragni, “Multispectral imaging approach in the diagnosis of
cutaneous melanoma: Potentiality and limits,” Phys. Med. Biol. 45,
1243–1254 (2000).
30
R. Levenson, P. J. Cronin, and K. K. Pankratov, “Spectral imaging for
brightfield microscopy,” Proc. SPIE 4959, 27–33 (2003).
31
H. Motomura, K. Ohsawa, N. Ohyama, M. Yamaguchi, and H. Haneishi,
“Natural Vision image data file format for spectral-based color
reproduction system,” Proc. IS&T’s PICs Conference (IS&T, Springfield,
VA, 2003) pp. 278–284.
32
H. Kanazawa, M. Yamaguchi, H. Haneishi, and N. Ohyama, “Natural
Vision XML Database,” Proc. AIC Colour 05–10th Congress of the
International Colour Association (ICA, Granada, Spain, 2005) pp.
1661–1664.
33
T. Uchiyama, M. Yamaguchi, H. Haneishi, N. Ohyama, and S. Nambu,
“A method for the unified representation of multispectral images with
different number of bands,” J. Imaging Sci. Technol. 48, 120–124 (2004).
34
M. Tsuchida, Y. Sakaguchi, H. Arai, M. Nishiko, N. Fujikawa, M.
Yamaguchi, H. Haneishi, and N. Ohyama, “High-fidelity color
reproduction based on multi-channel BTF/BRDF acquisition, rendering
and display,” ACM SIGGRAPH 2005 Sketches (ACM, Los Angeles, 2005)
p. 42.
35
T. Hyvarinen, E. Herrala, and A. Dall⬙Ava, “Direct sight imaging
spectrograph: A unique add-on component brings spectral imaging to
industrial applications,” Proc. SPIE 3302, 165–175 (1998).
36
A. Hirai, M. Hashimoto, K. Itoh, and Y. Ichioka, “Multichannel spectral
imaging system for measurements with the highest signal-to-noise
ratio,” Opt. Rev. 4, 334–341 (1997).
37
K. Ohsawa, T. Ajito, H. Fukuda, Y. Komiya, H. Haneishi, M. Yamaguchi,
and N. Ohyama, “Six-band HDTV camera system for spectrum-based
color reproduction,” J. Imaging Sci. Technol. 48, 85–92 (2004).
38
D. H. Marimont and B. A. Wandell, “Linear models of surface and
illuminant spectra,” J. Opt. Soc. Am. A 9, 1905–1913 (1992).
39
M. R. Pointer, “The gamut of real surface colours,” Color Res. Appl. 5,
145–155 (1980).
40
“Standard Object Colour Spectra Database for Colour Reproduction
Evaluation (SOCS),” TR X0012:98, Japanese Standards Association
(1998).
41
R. L. Alfvin and M. D. Fairchild, “Observer variability in metameric
color matches using color reproduction media,” Color Res. Appl. 22,
174–188 (1997).
42
K. Ohsawa, F. Konig, M. Yamaguchi, and N. Ohyama, “Color matching
experiment using 6-primary display,” 3rd International Conference on
Multispectral Color Science (Univ. Joensuu, Joensuu, Finland, 2001) pp.
85–88.
43
T. Ajito, K. Ohsawa, T. Obi, M. Yamaguchi, and N. Ohyama, “Color
conversion method for multiprimary display using matrix switching,”
Opt. Rev. 8, 191–197 (2001).
44
H. Motomura, “Color conversion for a multi-primary display using
linear interpolation on equi-luminance plane method (LIQUID),” J. Int.
SID 11, 371–378 (2003).
45
F. Konig, K. Ohsawa, M. Yamaguchi, N. Ohayama, and B. Hill, “A
multiprimary display: Optimized control values for displaying
tristimulus values,” Proc. IS&T’s PICS Conference (IS&T, Springfield, VA,
2002) pp. 215–220.
46
H. Kanazawa, “Color conversion for multiprimary displays using a
spherical average method,” Proc. IS&T/SID 12th Color Imaging
Conference (IS&T, Springfield, VA, 2004) pp. 65–69.
47
J. Takiue, S. Sugino, Y. Murakami, T. Obi, M. Yamaguchi, and N.
Ohyama, “Evaluation of smoothness in color gradation on multiprimary
display,” Proc. IS&T/SID 12th Color Imaging Conference (IS&T,
Springfield, VA, 2004) pp. 257–264.
48
T. Uchiyama, M. Yamaguchi, H. Haneishi, and N. Ohyama, “A visual
J. Imaging Sci. Technol. 52共1兲/Jan.-Feb. 2008
Yamaguchi, Haneishi, and Ohyama: Beyond RGB: Spectrum-based color imaging technology
evaluation of images reproduced on a multiprimary display by color
decomposition based on spectral approximation,” J. Imaging Sci.
Technol. 49, 410–417 (2005).
49
S. Yu, Y. Murakami, T. Obi, M. Yamaguchi, and N. Ohyama,
“Multispectral image compression for high fidelity colorimetric and
spectral reproduction,” J. Imaging Sci. Technol. 50, 64–72 (2006).
50
S. Takamura and Y. Yashima, “Multiband Video Signal Coding with
H.264/AVC. MPEG-4 Studio Profile and Motion JPEG 2000 and
Associated Signal Format,” J. Inst. Image Electronics Engineers Japan 33,
792–801 (2004).
51
R. Iwama, M. Mitsui, M. Yamaguchi, H. Haneishi, and N. Ohyama,
“Real-time multispectral and multiprimary video system,” Proc. AIC
Colour 05–10th Congress of the International Colour Association (ICA,
Granada, Spain, 2005) pp. 137–140.
52
T. Abe, Y. Murakami, M. Yamaguchi, N. Ohyama, and Y. Yagi, “Color
correction of pathological images based on dye amount quantification,”
Opt. Rev. 12, 293–300 (2005).
53
P. Bautista, T. Abe, M. Yamaguchi, Y. Yagi, and N. Ohyama, “Digital
staining for multispectral images of pathological tissue specimens based
on combined classification of spectral transmittance,” Comput. Med.
Imaging Graph. 29, 649–657 (2005).
54
M. Yamaguchi, M. Mitsui, Y. Murakami, H. Fukuda, N. Ohyama, and Y.
J. Imaging Sci. Technol. 52共1兲/Jan.-Feb. 2008
Kubota, “Multispectral color imaging for dermatology: Application in
inflammatory and immunologic diseases,” Proc. IS&T/SID 13th Color
Imaging Conference (IS&T, Springfield, VA, 2005) pp. 52–58.
55
M. Yamaguchi, R. Iwama, H. Kanazawa, N. Fujikawa, H. Fukuda, H.
Haneishi, N. Ohyama, H. Wada, T. Kambara, M. Aihara, Y. Yamakawa,
A. Nemoto, M. Furukawa, and Z. Ikezawa, “Color reproducibility of skin
lesions in multispectral video: Experimental evaluation,” Proc. IS&T/SID
14th Color Imaging Conference (IS&T, Springfield, VA, 2006) pp. 8–13.
56
H. Haneishi, T. Hasegawa, A. Hosoi, Y. Yokoyama, N. Tsumura, and Y.
Miyake, “System design for accurately estimating spectral reflectance of
art paintings,” Appl. Opt. 39, 6621–6632 (2000).
57
H. Fukuda, T. Uchiyama, H. Haneishi, M. Yamaguchi, and Nagaaki
Ohyama, “Development of 16-bands multispectral image archiving
system,” Proc. SPIE 5667, 136–145 (2005).
58
A. Uchida, K. Ohsawa, Y. Komiya, M. Yamaguchi, H. Haneishi, and N.
Ohyama, “Product photography for catalogs with multispectral camera,”
Proc. Spring Conference of the Japanese Society of Printing Science and
Technology (JSPST, Chiba, Japan, 2003) pp. 131–134 (in Japanese).
59
J. Kishimoto, M. Yamaguchi, H. Haneishi, and N. Ohyama,
“IRODORI—A Color-rich Palette Based on Natural Vision Technology,”
ACM SIGGRAPH 2004 Emerging Technologies (ACM, Los Angeles, 2004)
p. 13.
010201-15