The European hyperspectral imaging instrument OMEGA has been in operation around Mars since early 2004. OMEGA has constructed imaging maps covering almost the entire Martian surface (Bibring et al., 2005). The OMEGA science team have identified a 60 x 200km deposit rich in Ca-sulfates, most likely gypsum, near the Martian northern polar cap (Langevin et al., 2005). They have suggested an evaporation origin linked to varying Martian obliquity, warming the poles during periods of high obliquity, causing the melting of water ice, and then evaporation of this ephemeral water, leaving behind salty brines adjacent to the frozen ice cap. In 2004, a hyperspectral imaging survey of the Yilgarn Craton was carried out using the HyMap instrument (Cocks et al., 1998). The Yilgarn is a good Mars analog due to the presence of ultramafic-mafic volcanic basalt flows. We have therefore analysed this hyperspectral coverage of the evaporite deposits of the Yilgarn Craton to compare an accessible, Earth-based evaporate analog with the Mars sulfate deposits. Using standard hyperspectral analysis methods, we have mapped gypsum in Western Australian dry lake evaporite deposits near Kalgoorlie. In the future, we shall use an absorption band modelling method (Brown et. al., 2005) to determine the shape, position and intensity of absorption bands that are due to sulfates. Fieldwork in this area has enabled a laboratory based evaluation of hand samples of gypsum taken from the Western Australian deposits. Evaluation of this data and comparison to future Martian hyperspectral data will lead to a greater understanding of the hydration state and cation type of Martian sulfate deposits.

The possibility of using multispectral techniques to determine the concentration profiles of myoglobin derivatives as a function of the distance to the meat surface during meat oxygenation is demonstrated. Reduced myoglobin (Mb) oxygenated oxymyoglobin (MbO₂) and oxidized Metmyoglobin (MMb) concentration profiles are determined with a spatial resolutions better than of 0.01235 mm/pixel. Pigment concentrations are calculated using (K/S) ratios at isobestic points (474, 525, 572 and 610 nm) of the three forms of myoglobin pigments. This technique greatly improves previous methods, based on visual determination of pigment layers by their color, which allowed only estimations of pigment layer position and width. The multispectral technique avoids observer and illumination related bias in the pigment layer determination.

The skin's ability to retain moisture, which is hereafter referred as skin moisturizing-ability, is one of the important factors in skin health. Skin defends the biological tissue from the outside influences, skin sebum and moisture especially play an important role in that protection. The sebum and moisture meters available on the market, however, need to contact with skin. As a non-contact method, near-infrared (NIR) spectroscopic imaging, using absorption of the OH stretching overtone, has recently been capable of detecting changes in skin hydration of the forearms. However, face skin hydration has not been measured, and the moisture-related sebum has not been paid attention to, even though the face is important from the cosmetic and medical point of view. This study, therefore, aims to measure and visualize the spatial distribution of moisturizing-ability of the face skin by NIR spectroscopic imaging. The NIR spectral imaging system consists of two interference filters (1060 nm and 1450 nm) mounted on a filter wheel and a NIR camera with indiumgallium arsenide array sensor. We measured human face skins with/without moisturizing lotion and found that the glabella and nose have strong moisturizing-ability because of sebaceous glands. It was also shown that the areas where moisturizing lotion was applied were successfully displayed by subtracting two absorbance images measured at different wavelength bands. This technique can be applied to the functional assessment of face skin moisturizer in medicine and cosmetics.

A method is proposed for estimating the spectral reflectance of made-up skin color under various conditions including the undesirable colored skin. The color of dark spot is caused by increasing the component of melanin. The reddish skin is caused by the increase of hemoglobin. Our method uses the Kubelka-Munk theory to calculate the surface spectral reflectance human skin. This theory calculates the reflectance and transmittance of the light passing through a turbid medium from the absorption and scattering of the medium. The spectral reflectance of made-up skin is estimated by adjusting parameters of the thickness of the makeup layer. The proposed estimation method is evaluated on an experiment in detail. First, we measure the spectral reflectance of facial skin under the three conditions of normal skin, undesirable skin, and made-up skin. The undesirable skin includes stain, suntan or ruddy skin. The made-up skin means the skin with foundation on the normal skin, the stain, the suntan and the ruddy skin. Second, we estimate the spectral reflectance of made-up skins from the reflectance of bare skins and optical characteristics of foundations. Good coincidence between the estimated reflectance and the direct measurement shows the feasibility of the proposed method.

This document describes the principles of infrared thermography and its application to humanitarian demining in the world as well as the factors influencing its application in a country like Colombia which suffers badly the problem posed by antipersonnel mines. The main factors that affect the images taken by different sensors are: day time, mine size and material, installation angle, object's burial depth, moisture, emissivity, wind, rain, as well as other objects in the proximity shadowing the images. Infrared image processing methods and results of tests done in different sites of the country such as Cartagena, Bogota, and Tolemaida are also shown. Finally, a method for the detection of the presence of a buried object is presented with its successful results.

Moderate Resolution Imaging Spectroradiometer (MODIS) and Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) are multi-spectral sensors embarked on the EOS AM-1 (TERRA) satellite platform. Both sensors opperate in different spectral bands, but also with different pixel resolutions. The overall goal of this paper is to classify MODIS data to get an estimation of water surface area, very useful in the post-crisis periods for the decision makers at all levels. To develop the classification technique, the strategy was to obtain MODIS and ASTER data acquired at the same time over the same location, and use the ASTER data as "ground truth". Two lakes in the Bihor County of Romania were chosen and satellite data from October 31, 2002 were utilized. From the ASTER data we created a detailed water mask to be used as ground truth for the MODIS water classification. The percent water image derived from ASTER was superimposed on the MODIS image. A supervised classification for water was performed on the 3-band MODIS image using the feature space algorithm. The water surface area as measured from the MODIS classification was about 16% more than the ASTER ground truth-value. Due to the constraint that high spatial resolution satellite images are low temporal resolution, there exists a need for a reliable method to obtain accurate information from medium resolution data. This approach provided useful information concerning the water classification from different resolution data that could help in the estimation of water surface area from MODIS imagery.

Since commercial image detectors, such as charge-coupled device (CCD) cameras, have a limited dynamic range, it is difficult to obtain images that really are unsaturated, as a result of which the reflectance parameters may be inaccurately estimated. To solve this problem, we describe a method to estimate reflectance parameters from saturated spectral images. We separate reflection data into diffuse and specular components at 5-nm intervals between 380nm and 780nm for each pixel of the spectral images, which are captured at different incident angles, and estimate the diffuse reflectance parameters by applying the Lambertian model to the diffuse components. To estimate the specular reflectance parameters from the specular components, we transform the Torrance-Sparrow equation to a linear form, assuming Fresnel reflectance is constant. We then estimate specular parameters for intensity of the specular reflection and standard deviation of the Gaussian distribution, using the least squares method from unsaturated values of the specular components. Since Fresnel reflectance contributes to the physically based Torrance-Sparrow model in computer graphics and vision, we estimate both the Fresnel reflectance in terms of the Fresnel equation for the incident angle and the refractive index of the surface for dielectric materials, which varies with wavelength. We carried out experiments with measured data, and with simulated specular components at different saturation levels, generated according to the Torrance-Sparrow model. Our experimental results reveal that the diffuse and specular reflectance parameters are estimated with high quality.

The profit of low-cost, multispectral imaging systems in estimating spectral power distributions has been widely studied. There are various mathematical methods available (PCA, Wiener's estimation method, spline interpolation, MDST, among others) which permit the accurate reconstruction of a spectrum from the response of a small set of sensors. One important issue in this task is the influence of noise, its propagation through mathematical transformations and how the selection of the sensors of the multispectral system, combined with the spectral estimation algorithm chosen, may reduce its influence. We report here on four different spectral recovery methods that reconstruct skylight spectra from the responses of three Gaussian sensors (the spectral profile of which is a Gaussian curve). The sensors are searched for using a simulated annealing algorithm, and they are optimized so that they give the best possible spectral and colorimetric reconstructions, even in the presence of noise. We show here how the accuracy of the reconstructions is influenced by the recovery method chosen.

In recent times, the use of multispectral images has rapidly grown in various fields like defense, agriculture, medicine, etc. However, the image acquisition technology for multispectral images is still in its primitive stages compared to the technology used in commercial digital color cameras. Digital color cameras use mosaicked technology for acquiring and forming color images. An array of sensors is used to capture one spectral band per pixel location. The final image is then formed by filling the missing spectral band intensity values at each pixel location. This process of estimating the full color image from the acquired sensor data is called demosaicking. In this paper, we propose to use the mosaicked technology for multispectral image acquisition systems. This paper focuses on developing demosaicking methods for such multispectral image acquisition systems. We explore ways of extending the existing demosaicking methods to multispectral images. This paper also addresses the problem of noise and degradations present during the acquisition process. The existing demosaicking methods tend to fail in the presence of external noise and degradations. To solve this problem, we have developed a maximum aposteriori probability (MAP) based method that performs demosaicking and at the same time reduces noise and degradations in the output. This novel approach treats the demosaicking problem as an image restoration problem and solves the optimization problem using the gradient descent method. The experimental results show that the MAP based demosaicking method provides a superior output compared to the traditional demosaicking methods. Various performance metrics have been used to compare results from different demosaicking algorithms.

The noise present in a color image acquisition system influences the accuracy of the estimated colorimetric values and the accuracy of the recovered spectral reflectances of objects being imaged through the use of sensor responses. Estimation of the noise levels in the devices is important for the accurate acquisition of colorimetric or spectral information. This work addresses the problem for the determination of noise variances in multispectral image acquisition systems. Several models for the determination are compared and experimental results to show the accuracy of the model proposed by the author are demonstrated. It is shown that the estimates by the proposal agree fairy well with the noise variance which minimizes the mean square errors (MSE) of the recovered spectral reflectances by the use of the Wiener filter.

Multi-spectral laser imaging can be a useful technology for target discrimination, classification, and identification based on object spectral signatures. The mid-IR region (~3-14 μm) is particularly rich of molecular spectroscopic fingerprints, but the technology has been under utilized. Compact, potentially inexpensive semiconductor lasers may allow more cost-effective applications. This paper describes a development of semiconductor-laser-based multi-spectral imaging for both near-IR and mid-IR, and demonstrates the potential of this technology. The near-IR study employed 7 wavelengths from 0.635-1.55 μm, and used for system engineering evaluation as well as for studying the fundamental aspects of multi-spectral laser imaging. These include issues of wavelength-dependence scattering as a function of incident and receiving angle and the polarization effects. Stokes vector imaging and degree-of-linear-polarization were shown to reveal significant information to characterize the targets. The mid-IR study employed 4 wavelengths from 3.3-9.6 μm, and was applied to diverse targets that consist of natural and man-made materials and household objects. It was shown capable to resolve and distinguish small spectral differences among various targets, thanks to the laser radiometric and spectral accuracy. Colorless objects in the visible were shown with "colorful" signatures in the mid-IR. An essential feature of the study is an advanced system architecture that employs wavelength-division-multiplexed laser beams for high spectral fidelity and resolution. In addition, unlike conventional one-transmitter and one receiver design, the system is based on a scalable CDMA network concept with multiple transmitters and receivers to allow efficient information acquisition. The results suggest that multi-spectral laser imaging in general can be a unique and powerful technology for wide ranging applications.

In this paper, we propose a flat-bed scanner system to record spectral and glossiness information for various sheets like objects. In the proposed system, five filters are used to acquire the multi-band images of the object. The spectral reflectance image can be estimated and be recorded from the multi-band images. The glossiness of the object is recorded as two images taken by the different geometries about illuminant which is from 45 degrees, and from 0+α degrees respectively when the averaged normal vector of the sheet like object is defined as 0 degree. We performed two types of computer simulation by using the two images to reproduce various appearance of recorded object. As the first simulation, the various appearances of image are reproduced through a weighted linear combination of the two images. As the second simulation, the normal vector distribution of the object is estimated from the image taken by 45 degree illuminant. By using this normal vector distribution, roughness of the object is estimated from the image taken by 0 + α degree illuminant. The normal vector and estimated roughness are used to reproduce the various appearance of the object under arbitrary illuminant.

This work is focused on the study and comparison of the performance for color measurements of different systems based on optoelectronic imaging sensors. We used two different configurations of the imaging system, one with three acquisition channels and the other with more spectral bands, in order to measure the color associated to each pixel of the captured scene. We applied different methodologies to obtain the XYZ tristumulus values from the measured digital signals. The different techniques included an absolute spectral and colorimetric characterization of the system and also direct transformations between both sets, which used several mathematical fittings such as the pseudo-inverse technique, a non-linear estimation method and the principal component analysis. The proposed configurations were experimentally tested imaging the patches of the Gretagmacbeth ColorChecker DC and Color Rendition charts placed in a light booth, and measuring the corresponding colors. The results obtained showed that optoelectronic imaging systems can be used in order to perform rather accurate color measurements with high spatial resolution. Specifically, the best results in terms of CIELab color differences were achieved by using a multispectral configuration of the imaging system with seven spectral bands and directly transforming the digital signals into XYZ tristimulus values by means of the pseudo-inverse technique.

In addition to the great advancement of high-resolution and large-screen imaging technology, the issue of color is now receiving considerable attention as another aspect than the image resolution. It is difficult to reproduce the original color of subject in conventional imaging systems, and that obstructs the applications of visual communication systems in telemedicine, electronic commerce, and digital museum. To breakthrough the limitation of conventional RGB 3-primary systems, "Natural Vision" project aims at an innovative video and still-image communication technology with high-fidelity color reproduction capability, based on spectral information. This paper summarizes the results of NV project including the development of multispectral and multiprimary imaging technologies and the experimental investigations on the applications to medicine, digital archives, electronic commerce, and computer graphics.

In this paper, the encoding of spectra is studied considering a variety of different observers on the one hand and a large set of standardized and non-standardized illuminants on the other. A number of 24 different observers has been defined including the CIE 1931 standard observer and the CIE 1964 supplementary standard observer as well as standard deviators. Others are selected with respect to largest differences from measurements published by Stiles and Burch. In addition, different illuminants are applied. Altogether, 52 illuminants are considered for reproducing color stimuli including such as D ₅₀, D₅₅, D₆₅, D₇₅, A, B, C, E, and F₁ to F₁₂ as well as measured ones. So, this set contains illuminants with uniform spectral radiating power on the one hand and spiky power distribution on the other. A number of different encoding methods has been tested by calculating color encoding errors for all combinations of observers, illuminants and a representative set of test spectra assembled from data of Vrhel and Pointer. The encoding methods considered are based on expansions into basis functions derived from original or pre-distorted test spectra, Fourier and sine series as well as expansions into modifications of the latter ones. Maximum as well as average errors are presented.

In this paper, we propose a new color matching method between a proof and a target print by using a projection display system based on the spectral color reproduction. In this method, a color of proof is corrected by synthesizing a projection image which is calculated to minimize the color difference between each print. The radiance of the proof and the target print are calculated by using the reflectance of the prints and the radiance from the projector, and we use the method based on the XYZ tristimulus values (colorimetric method) and the spectral values (spectral-based method). We compared the color difference between the colorimetric method and spectral-based method. The average color difference ΔE*₉₄ by using the colorimetric method was 4.00. On the other hand, the color difference used the spectralbased method was 2.13. From these results, we concluded that the spectral-based method is more effective than the colorimetric method to perform the accurate color reproduction by synthesizing the projection color and the proof.

In this paper, we present results for optimizing images for an industrial show room. The light conditions are not very controllable and the projector is not a high quality one. The optimization is done using metameric reproduction and to do this we measure spectral information of the product, projector and the illumination at the show room. The spectral characteristic of the red channel of the projector was surprising: the range of possible red values was narrower than the green and blue range. This caused some limitations which needed to be taken into account in calculating the optimal images: optimal images can have either full contrast range with a reddish tint or correct hue with narrower contrast range.

All imaging devices have two gamuts: the stimulus gamut and the response gamut. The response gamut of a print engine is typically described in CIE colorimetry units, a system derived to quantify human color response. More fundamental than colorimetric gamuts are spectral gamuts, based on radiance, reflectance or transmittance units. Spectral gamuts depend on the physics of light or on how materials interact with light and do not involve the human's photoreceptor integration or brain processing. Methods for visualizing a spectral gamut raise challenges as do considerations of how to utilize such a data-set for producing superior color reproductions. Recent work has described a transformation of spectra reduced to 6-dimensions called LabPQR. LabPQR was designed as a hybrid space with three explicit colorimetric axes and three additional spectral reconstruction axes. In this paper spectral gamuts are discussed making use of LabPQR. Also, spectral gamut mapping is considered in light of the colorimetric-spectral duality of the LabPQR space.

The spectral reflectance of icons is measured using a measurement system developed in our previous study, and it is applied to detect metameric color areas in the icons. In this paper, a technique for detecting metameric color areas is proposed and examined by using a test chart and ten icons painted on wooden plates. In the proposed technique, a coefficient showing the degree of metamerism is proposed; based on the definition of metamerism whereby two stimuli can match in color while having different spectral reflectance functions. The experimental results can then be used to consider which parts of the icons have previously been repainted as restoration treatments. Despite the necessity of further consideration using certain chemical analyses and so on to conclude whether or not the experimental results are reliable, they demonstrate that the proposed technique has the basic ability to detect metameric color areas.

We present a scanning device for 32-band multi-spectral imaging of paintings in the 380÷800 nm spectral region. The system is based on contact-less and single-point measurement of the spectral reflectance factor. Multi-spectral images are obtained by scanning the painted surface under investigation. An adjustment procedure was established and calibration was performed by means of a set of seven matt ceramic color tiles certified by National Physical Laboratory (UK). Colorimetric calculations were carried out in the XYZ colorimetric space, by following the CIE recommendations and choosing the D65 standard illuminant and the 1931 standard observer. Measurement campaigns were carried out on several paintings in situ and at the INOA Optical Metrology Laboratory located inside the Opificio delle Pietre Dure in Florence. As an example we report herein on the measurements carried out on the Madonna in gloria tra Santi by Andrea Mantegna, (at present in the Pinacoteque of the Castello Sforzesco in Milan). Multivariate image analyses (MIA) were performed by considering the multi-spectral images as three-way data set. The stack of detected images were unfolded in a 2D data matrix and analyzed by the conventional Principal Component Analysis (PCA).

The multiband video camera and display system has developed to increace quality of the video image. The multiband means that there has more than three color channels. The new applications can be created by increasing the number of camera primaries, for example an influence of illumination can be fixed accurately. Furthermore, increasing number of primaries gives a good possibility for image and video processing. However, the growed amount of data causes problems for data transform, store, and process. In this paper, we introduce six band video capturing system and our spectral video database. In the database there is a lot of different type of video clips and the size of spectral video database is more than 300 giga bytes (GB). We also have developed compression scheme for spectral video based on principal component analysis (PCA) and JPEG2000 methods. Here, we concentrate to compress spectral video sequence frame by frame.

The focus of this paper is on the broad extension of multi-spectral color to both scientist and consumer by creating camera/projector arrays composed of commodity hardware. In contrast to expensive, high-maintainance systems which rely on the physical registration of device spaces, we rely on the virtual alignment of viewing spaces in software where real-time alignment is achieved using the processing capacity of the graphical processing units of consumer PC video cards. Specifically, this paper focuses on the inclusion of real-time, composite pattern, structured light illumination (SLI) as a means of recording the 3D shape of objects, which will then be used for the registration of single-color images taken from multiple view points simultaneously. As such, the described system is able to achieve a cost per unit that scales linearly with the number of color primaries.

Device gamuts are commonly defined for output devices, such as monitors or printers. In this paper, a definition of gamuts of input devices will be examined, considering multispectral cameras as examples. A method appropriate to calculate them as a function of the camera model and the spectral reconstruction algorithm will be proposed. The method will be applied to multispectral camera models with a variable number of channels. The characteristics of the resulting gamuts will be shown and examined as a function of the number of channels. Implications on the minimum number of channels needed will be derived. The method proposed here to characterize input devices can be used in addition to common quality criteria such as color distances like ΔE₀₀, spectral errors, etc. The advantage of the proposed method is the independence of any given spectral data set. This makes it a quality criterion universal for linear (multispectral) cameras and reconstruction algorithms.

A new, semantically meaningful technique for querying the images from a spectral image database is proposed. The technique is based on the use of both color- and texture features. The color features are calculated from spectral images by using the Self-Organizing Map (SOM) when methods of Gray Level Co-occurrence Matrix (GLCM) and Local Binary Pattern (LBP) are used for constructing the texture features. The importance of texture features in a querying is seen in experimental results, which are given by using a real spectral image database. Also the differences between the results gained by the use of co-occurrence matrix and LBP are introduced.