Roland Harig

Noninvasive Analysis of Paintings by Mid‐infrared Hyperspectral Imaging

In recent years, in situ noninvasive spectroscopy has played an increasingly important role in art conservation. Spectroscopic methods can be used to gain a deep understanding of the material composition of art objects while fully respecting their integrity and value. Considerable improvements in detector technology, instrument–computer interfacing, focusing optics, and the performance of radiation sources have been made in the spectroscopic field and have led to the development and successful application of a series of noninvasive and portable analytical tools for point examination. In this context, current scientific interest is focused on the development of mapping/imaging multi-/hyperspectral methods, since area examination naturally meets the demands of a holistic art-historical approach by revealing not only the chemical composition of painting materials but also their semiquantitative spatial distribution with respect to what is visible to the naked eye. Recently, the possibility of mapping elemental distribution on paintings by means of a portable scanning macro X-ray fluorescence device was demonstrated to be useful for the investigation of the materials used by artists. The molecular identification and spatial distribution of a number of pigments can be inferred from reflection imaging in the visible (Vis: 400–750 nm) and near-infrared regions (NIR: 750–2500 nm) by combining information on electronic transitions in the visible range with overtone and combination vibrational bands in the near-infrared range which are associated with inorganic pigments containing hydroxy groups, such as lead white, azurite, and gypsum. The identification and mapping of organic compounds is a challenge for the noninvasive analysis of artworks. In comparison with inorganic pigments and fillers, there is a larger variety of organic compounds, and they are present in smaller amounts and more subject to chemical alteration. If both traditional and modern art are taken into consideration, just some of the organic compounds to be identified and localized may include: drying oils, proteins, acrylics, alkyds, polyvinyl acetates, natural and synthetic waxes, terpene resins, and natural and synthetic dyes. Recently, Ricciardi et al. demonstrated the potential of hyperspectral imaging in the NIR region (10000–4000 cm ) to discriminate between lipid and proteinaceous binders on illuminated manuscripts on the basis of the vibrational combination/overtone bands of methylene groups and amides. In contrast, the mid-infrared (MIR) range has not yet been exploited for the spatially resolved remote study of artworks, although the MIR range (4000–600 cm ) has proved to be very informative for the identification of artists materials by noninvasive point analysis 12] or microinvasively by cross-section imaging. Herein, we describe the potential of imaging MIR spectroscopy for painting analysis through the application of a novel hyperspectral imaging system (HI90, Bruker Optics). The system developed for the remote identification and mapping of hazardous compounds was adapted in this study for reflection measurements of paintings by the use of an external infrared radiation source. The subject of our study was a painting by Alberto Burri, Sestante 10 (1982), which is currently exhibited at the Ex-Seccatoi del Tabacco (Perugia, Italy). Three areas (ca. 9 9 cm) of the large painting (250 360 cm) were analyzed on site with the HI90 system. We also investigated several points within the same areas noninvasively with a portable FTIR spectrometer (Alpha-R, Bruker Optics) to validate the assignment made on the basis of the HI90 spectral data. In Figure 1a, the visible image of the investigated area I of Sestante 10 is shown. The resulting brightness temperature difference images in Figure 1b,c clearly highlight the use of a different binder for the orange sector to that used for the other sectors. More precisely, the image depicted in Figure 1b shows the false-color representation of the difference in the mean brightness temperature for a peak in the signature of the orange area (1154–1167 cm ) and the mean brightness temperature for a frequency range in which the reflectance is lower (1197–1209 cm ; see range (b) in Figure 1d). The area in the brightness temperature image in Figure 1b indicates the presence of an acrylic binder, as suggested by a comparison with the spectrum recorded with the HI90 instrument [*] Dr. F. Rosi, Dr. C. Miliani Istituto di Scienze e Tecnologie Molecolari CNR-ISTM, SMAArt Via Elce di sotto, 9 Perugia 06123 (Italy) E-mail: costanza.miliani@cnr.it

Remote sensing of atmospheric pollution by passive FTIR spectrometry

Passive remote sensing with a Fourier transform IR (FTIR) spectrometer allows the detection and identification of pollutant clouds in the atmosphere. In this work the measurement technique and a data analysis method that does not require a previously measured background spectrum are described. Recent experimental results obtained with anew high sensitive FTIR remote sensor are presented. Many situations do not allow the measurement of a background spectrum prior to the measurement of pollutants in order to perform background removal. After a radiometric calibration of the FTIR spectrometer with IR reference sources the spectral radiance of the environment can be measured. With the inverse function of Plancks radiation law the brightness temperature is computed. The temperature spectrum has a constant baseline for many natural materials that serve as the background in field measurements because their emittance is high and almost constant in the spectral range 800-1200 cm-2. The influence of environmental and instrumental parameters on the sensitivity of the method are discussed. Experimental results are presented to illustrate the enhancement of the signal to noise ratio that can be achieved by the alignment of the spectrometer to backgrounds with a high temperature difference to the environment.

Passive remote sensing of pollutant clouds by Fourier-transform infrared spectrometry: signal-to-noise ratio as a function of spectral resolution

In a passive infrared remote sensing measurement, the spectral radiance difference caused by the presence of a pollutant cloud is proportional to the difference between the temperature of the cloud and the brightness temperature of the background (first-order approximation). In many cases, this difference is of the order of a few kelvins. Thus the measured signals are small, and the signal-to-noise ratio (SNR) is one of the most important quantities to be optimized in passive remote sensing. A model for the SNR resulting from passive remote sensing measurements with a Fourier-transform infrared spectrometer is presented. Analytical expressions for the SNR of a single Lorentzian line for the limiting cases of high and low spectral resolutions are derived. For constant measurement time the SNR increases with decreasing spectral resolution, i.e., low spectral resolutions yield the highest SNRs. For a single scan of the interferometer, a spectral resolution that maximizes the SNR exists. The calculated SNRs are in good agreement with the measured SNRs.

Remote sensing of gases by hyperspectral imaging: system performance and measurements

Abstract. Remote gas detection and visualization provides vital information in scenarios involving gas leaks, environmental monitoring, chemical accidents or attacks. Imaging systems based on Fourier transform spectrometers with single detector elements have been applied for several years by emergency response forces for gas identification and quantification. In this work a hyperspectral imager employing a Michelson interferometer and an infrared focal plane array detector is characterized. The system provides spatially resolved spectral information about the measurement scene. The performance of the system is evaluated by laboratory measurements. Results of gas detection in the field are presented and discussed. The gas detection algorithm is based on a physical model for the measured radiance. In this model the atmosphere is divided into multiple homogenous layers of constant temperature. The signatures of the gases present in these layers are then compensated in the measured spectrum. No information about the signature of the background is required. Moreover an algorithm that combines spectral and spatial information is presented. This algorithm enhances the signal to noise ratio of the spectra and thus improves the detection limits. Using these algorithms it is possible to identify, visualize, and track gas clouds in real time.

Scanning infrared remote sensing system for identification, visualization, and quantification of airborne pollutants

Remote sensing by Fourier-transform infrared (FTIR) spectrometry allows detection, identification, and quantification of airborne pollutants. In the case of leaks in pipelines or leaks in chemical plants, chemical accidents, terrorism, or war, hazardous compounds are often released into the atmosphere. Various Fourier-transform infrared spectrometers have been developed for the remote detection and identification of hazardous clouds. However, for the localization of a leak and a complete assessment of the situation in the case of the release of a hazardous cloud, information about the position and the size of a cloud is essential. Therefore, an imaging passive remote sensing system comprised of an interferometer (Bruker OPAG 22), a data acquisition, processing, and control system with a digital signal processor (FTIR DSP), an azimuth-elevation-scanning mirror, a video system with a DSP, and a personal computer has been developed. The FTIR DSP system controls the scanning mirror, collects the interferograms, and performs the Fourier transformation. The spectra are transferred to a personal computer and analyzed by a real-time identification algorithm that does not require background spectra for the analysis. The results are visualized by a video image, overlaid by false color images. For each target compound of a spectral library, images of the coefficient of correlation, the signal to noise ratio, the brightness temperature of the background, the difference between the temperature of the ambient air and the brightness temperature of the background, and the noise equivalent column density are produced. The column densities of all directions in which a target compound has been identified may be retrieved by a nonlinear least squares fitting algorithm and an additional false color image is displayed. The system has a high selectivity, low noise equivalent spectral radiance, and it allows identification, visualization, and quantification of pollutant clouds.

Short-range remote detection of liquid surface contamination by active imaging Fourier transform spectrometry

An imaging Fourier transform spectrometer developed at TUHH was used for short-range remote detection and identification of liquids on surfaces. The method is based on the measurement of infrared radiation emitted and reflected by the surface and the liquid. A radiative transfer model that takes both the real and imaginary parts of the refractive index of the materials into account has been developed. The model is applied for the detection and identification of potentially hazardous liquids. Measurements of various liquids on diverse surfaces were performed. The measured spectra depend on the optical properties of the background surface. However, using the radiative transfer model, automatic remote detection and identification of the liquids is possible. The agreement between measured spectra and spectra calculated using the radiative transfer model is excellent.

Infrared remote sensing of hazardous vapours: surveillance of public areas during the FIFA Football World Cup 2006

The German ministry of the interior, represented by the civil defence agency BBK, established analytical task forces for the analysis of released chemicals in the case of fires, chemical accidents, terrorist attacks, or war. One of the first assignments of the task forces was the provision of analytical services during the football world cup 2006. One part of the equipment of these emergency response forces is a remote sensing system that allows identification and visualisation of hazardous clouds from long distances, the scanning infrared gas imaging system SIGIS 2. The system is based on an interferometer with a single detector element in combination with a telescope and a synchronised scanning mirror. The system allows 360° surveillance. The system is equipped with a video camera and the results of the analyses of the spectra are displayed by an overlay of a false colour image on the video image. This allows a simple evaluation of the position and the size of a cloud. The system was deployed for surveillance of stadiums and public viewing areas, where large crowds watched the games. Although no intentional or accidental releases of hazardous gases occurred in the stadiums and in the public viewing areas, the systems identified and located various foreign gases in the air.

3-D Reconstruction of Gas Clouds by Scanning Imaging IR Spectroscopy and Tomography

In the case of accidents at chemical plants, during transportation of chemicals, or after terrorist attacks, hazardous compounds may be released into the atmosphere. The weather-dependent propagation of these toxic clouds can threaten population and environment. In order to apply appropriate safety measures, it is necessary for emergency response forces to detect and identify the regarding substances. In addition, it is important to determine position, dimensions, and source of the gas cloud. Moreover, it is desirable to perform the necessary measurements from a distance to minimize the threat for emergency response personnel. Imaging remote sensing by IR spectroscopy provides a method for generating (2-D) images of the cloud. Combined with an appropriate visible (video) or IR image of the scene, these images can reveal information like the dimensions and the location of the source of the cloud. Nevertheless, the distance between the system and the cloud and the dimensions of the cloud along the line of sight are not available if a single image is measured. If images of the cloud are recorded from at least two different positions at the same time, information about the position and the 3-D shape of the cloud becomes available. Therefore, a method for 3-D reconstruction of gas clouds based on imaging IR spectroscopy and tomography has been developed. The remote sensing system, the measurement setup, and the algorithm generating the 3-D structures from the images are described. The method is applied to reconstruct the exhaust gas plume of an industrial stack.

New scanning infrared gas imaging system (SIGIS 2) for emergency response forces

The German ministry of the interior, represented by the civil defense agency BBK, is currently establishing analytical task forces for the analysis of released chemicals in the case of fires or chemical accidents. One part of the equipment of these emergency response forces will be a remote sensing system that allows the identification of hazardous clouds from long distances. Therefore, a new scanning infrared gas imaging system, SIGIS 2, is currently being developed at TUHH. The system is based on an interferometer with a single detector element (Bruker OPAG 33) in combination with a telescope and a synchronized scanning mirror. The new scanning system allows 360° surveillance. For simple interpretation of the results, the system is equipped with a video camera and the results of the analyses of the spectra are displayed by an overlay of a false color image on the video image. This allows a simple evaluation of the position and the size of a cloud. In order to allow simultaneous display of false color representations of measurement results and of the video image in real-time, a new scanner module has been developed. In the standard measurement mode, 16 two-sided interferograms per second are measured, analyzed, and the results are displayed. The spectral resolution is 4 cm-1. The new interferometer, the new scanning system, the data analysis method, and first results of measurements are presented.

Optical Remote Sensing for Characterizing the Spatial Distribution of Stack Emissions

In this contribution, optical methods based on passive FTIR (Fourier Transform Infrared) and DOAS (Differential Optical Absorption Spectroscopy) techniques have been used to characterize the dispersion of gas emissions from industrial sources. Portable, zenith-looking, passive-DOAS instruments measured the horizontal distribution of an SO 2 plume from a power plant in a coastal town of Mexico. The column density of this gas was measured while making traversals across the plume with a car and a boat downwind from the emission source. The cross sections measured at different distances from the source are used to characterize the horizontal dispersion and to estimate emission fl uxes. In addition, a Scanning Infrared Gas Imaging System (SIGIS) was used to acquire passive IR spectra at 4 cm resolution in a two-dimensional array, from which a false-color image is produced representing the degree of correlation of a specifi c gaseous pollutant. The 24-h, real-time animations of the SO 2 plume help us to understand dispersion phenomena in various atmospheric conditions. The wealth of information retrieved from these optical remote sensors provides an alterative method for evaluating the results from plume dispersion models.