Multispectral Thermal Imager Research Articles

Development of Multispectral Thermal Imager for Measuring the Temperature of Long-distance Explosion Flames

Development of Multispectral Thermal Imager for Measuring the Temperature of Long-distance Explosion Flames

Development of a multi-spectral thermal imager for measurement of the laser-induced damage temperature field

In order to analyze the damage effectiveness of the high-energy laser beam damage to the target, it is important to accurately measure the laser-induced damage temperature field. However, the obtained data cannot meet the requirements of damage effectiveness evaluation, as most of these data are from simulated temperature measurements in the laser-induced damage temperature field. Moreover, traditional temperature measuring instruments can only measure temperature points, and cannot obtain the multi-spectral temperature field of the laser-induced damage. In response to the above problems, this paper has developed a multi-spectral thermal imager for measurement of the laser-induced damage temperature field. Using a five-channel four-wavelength filter photometer, the instrument makes the fifth channel as a calibration channel so that the infrared thermal imager can accurately collect data under multi-spectral channels. Temperature fields corresponding to the spectral data can be inverted by multi-spectral thermometry method. The temperature results of the multi-spectral thermal imager are compared with that of thermocouples, and the measurement error is<1.5%. According to the experimental data of laser-induced damage field measurements, it is shown that the laser-induced damage temperature field can be acceptably measured with the multi-spectral thermal imager.

@thermogramer: Thermal Imaging as a Tool for Science Communication and E-Learning in Social Media

The COVID-19 pandemic boosted the presence of thermal cameras in our society. These devices are becoming cheaper and smaller and can even be plugged in our smartphones. Therefore, soon enough everybody will have access to these instruments. Thermal cameras have been widely used for industrial, research and/or academic purposes. Now, in the rise of the online era, this work proposes and assesses a new application for such devices as visual engaging tools for science communication and e-learning in social media. Here, we introduce @thermogramer as a science communication channel that shows multispectral (optical and thermal) images of daily life objects to explain the science behind different topics of social interest (climate change, emerging technologies, health, and popular traditions). This young project is already present in social media, press, TV and museum’s exhibitions, and its designed content have been already useful for new inexperienced users, science educators and communicators.

Reversible optical tuning of GeSbTe phase-change metasurface spectral filters for mid-wave infrared imaging.

Tunable narrowband spectral filtering across arbitrary optical wavebands is highly desirable in a plethora of applications, from chemical sensing and hyperspectral imaging to infrared astronomy. Yet, the ability to reconfigure the optical properties, with full reversibility, of a solid-state large-area narrowband filter remains elusive. Existing solutions require either moving parts, have slow response times, or provide limited spectral coverage. Here, we demonstrate a 1-inch diameter continuously tunable, fully reversible, all-solid-state, narrowband phase-change metasurface filter based on a GeSbTe-225 (GST)-embedded plasmonic nanohole array. The passband of the presented device is with transmittance and operates across the 3–5 µm thermal imaging waveband. Continuous, reconfigurable tuning is achieved by exploiting intermediate GST phases via optical switching with a single nanosecond laser pulse, and material stability is verified through multiple switching cycles. We further demonstrate multispectral thermal imaging in the mid-wave infrared using our active phase-change metasurfaces. Our results pave the way for highly functional, reduced power, compact hyperspectral imaging systems and customizable optical filters for real-world system integration.

Tunable mid-wave infrared Fabry-Perot bandpass filters using phase-change GeSbTe.

We demonstrate spectrally-tunable Fabry-Perot bandpass filters operating across the MWIR by utilizing the phase-change material GeSbTe (GST) as a tunable cavity medium between two (Ge:Si) distributed Bragg reflectors. The induced refractive index modulation of GST increases the cavity’s optical path length, red-shifting the passband. Our filters have spectral-tunability of ∼300 nm, transmission efficiencies of 60-75% and narrowband FWHMs of 50-65 nm (Q-factor ∼70-90). We further show multispectral thermal imaging and gas sensing. By matching the filter’s initial passband to a CO2 vibrational-absorption mode (∼4.25 µm), tunable atmospheric CO2 sensing and dynamic plume visualization of added CO2 is realized.

MEMS-Based Tunable Fabry–Perot Filters for Adaptive Multispectral Thermal Imaging

This paper reports on a proof-of-concept micro-electromechanical system-based Fabry–Perot filter that is capable of electrically tuning within the long-wave infrared thermal imaging band of 8–12 $\mu \text{m}$ . The device employs a single-layer quarter-wavelength thick tensile germanium membrane for the suspended top mirror in order to achieve nanometer-scale as-released mirror flatness across an area of several hundred square micrometers without any extraneous stress management techniques. Mechanical and optical characterization of the tunable filters of various sizes are presented and compared. A 200- $\mu \text{m}$ dimension square filter is demonstrated with $\mu \text{m}$ , namely, peak transmission above 80%, full-width at half-maximum of spectral passband of approximately 500 nm, and out-of-band rejection greater than 40:1. Optical modeling shows that this filter can achieve a pixel-to-pixel transmission peak wavelength variation of less than 1.2% across the entire $200~\mu \text{m}~ \times200$ - $\mu \text{m}$ optical imaging area. These results exceed the optical performance requirements for passive multispectral thermal imaging applications based on large-area focal plane arrays. In comparison, the 500 and 1000- $\mu \text{m}$ dimension filters are shown to exhibit significant mirror bowing with actuation and, thus, for a pixel-to-pixel transmission peak wavelength non-uniformity of $11.3~\mu \text{m}$ , respectively. [2015-0252]

A Switchable Mid-Infrared Plasmonic Perfect Absorber with Multispectral Thermal Imaging Capability.

A switchable perfect absorber with multispectral thermal imaging capability is presented. Aluminum nanoantenna arrays above a germanium antimony telluride (GST) spacer layer and aluminum mirror provide efficient wavelength-tunable absorption in the mid-infrared. Utilizing the amorphous-to-crystalline phase transition in GST, this device offers switchable absorption with strong reflectance contrast at resonance and large phase-change-induced spectral shifts.

Inverting multispectral thermal time-series images of volcanic eruptions for lava emplacement models

Abstract We present a novel method for interpreting time series of multispectral observations of volcanic eruptions. We show how existing models relating to radiance and area emplacement can be generalized into an integration-convolution of a Net Area Emplacement (NAE) function and a cooling function, assuming all surfaces follow the same cooling curve. The NAE describes the variation in the rate of emplacement of hot material with time and temperature, while the cooling function describes the cooling of a hot surface with time. Discretizing the integration-convolution equation yields an underdetermined matrix equation that we solve using second-order Tikhonov regularization to stabilize the solution. We test the inversion by modelling plausible NAE surfaces, calculating the radiances, adding noise and inverting for the original surface. Three or more spectral bands are required to capture the overall shape of the NAE, and recovering specific quantities is difficult. Single wavebands that yield flat kernels recover the total area emplacement curve (rate of increase of hot area – the integral of the NAE with respect to temperature) surprisingly well due to their property of conserving NAE, suggesting novel methods for calculating area emplacement rates (and effusion rates) from time series of satellite images and radiometer measurements.

Self-Adaptive Gradient-Based Thresholding Method for Coal Fire Detection Based on ASTER Data—Part 2, Validation and Sensitivity Analysis

The self-adaptive gradient-based thresholding (SAGBT) method is a simple non-interactive coal fire detection approach involving segmentation and a threshold identification algorithm that adapts to the spatial distribution of thermal features over a landscape. SAGBT detects coal fire using multispectral thermal images acquired by the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) sensor. The method was detailed by our previous work “Self-Adaptive Gradient-Based Thresholding Method for Coal Fire Detection Based on ASTER Data—Part 1, Methodology”. The current study evaluates the performance of SAGBT and validates its results by using ASTER thermal infrared (TIR) images and ground temperature data collected at the Wuda coalfield (China) during satellite overpass. We further analyzed algorithm performance by using nighttime TIR images and images from different seasons. SAGBT-derived fires matched fire spots measured in the field with an average offset of 32.44 m and a matching rate of 70%–85%. Coal fire areas from TIR images generally agreed with coal-related anomalies from visible-near infrared (VNIR) images. Further, high-temperature pixels in the ASTER image matched observed coal fire areas, including the major extreme high-temperature regions derived from field samples. Finally, coal fires detected by daytime and by nighttime images were found to have similar spatial distributions, although fires differ in shape and size. Results included the stratification of our study site into two temperature groups (high and low temperature), using a fire boundary. We conclude that SAGBT can be successfully used for coal fire detection and analysis at our study site.

A technique for mapping Fe/Mg-rich phyllosilicates on Mars using THEMIS multispectral thermal infrared images

[1] We present a method for mapping exposures of Fe/Mg-rich phyllosilicates on the Martian surface using THEMIS daytime multispectral thermal infrared images. The primary utility of this technique comes from the fact that THEMIS coverage is near-global at 100 m/pixel, so it offers the opportunity to extend and improve upon phyllosilicate maps created using CRISM and/or OMEGA near-infrared data alone. The spectral index formulation used for this mapping consists of a sum of spectral band ratios, weighted in different proportions. An initial formulation of the index was determined using a suite of synthetic spectra containing varying proportions of a Martian basalt spectrum and a laboratory phyllosilicate spectrum. A scene-specific formulation was then optimized by exploiting the correlation between pixels the initial formulation flags in each THEMIS scene as containing phyllosilicates and those identified in overlapping CRISM data. Once the optimal formulation of the index is determined, it is applied to the entire THEMIS scene, extending the map of phyllosilicates beyond the area of overlapping CRISM coverage. The method is illustrated using 12 THEMIS scenes from the Mawrth Vallis and Nili Fossae regions. The new THEMIS-derived maps of Fe/Mg-rich phyllosilicates in these two regions have approximately double the coverage provided by CRISM data. The cross-wavelength and cross-instrument approach we use to derive phyllosilicate indices for THEMIS data could potentially be used to define spectral indices for mapping other compositions of interest on Mars that possess spectral features in more than one spectral region.

Wavelength-tunable microbolometers with metamaterial absorbers

Microbolometers are modified by metallic resonant absorber elements, leading to an enhanced responsivity at selectable wavelengths. The dissipative energy absorption of tailored metamaterials allows for engineering the response of conventional bolometer microbridges. The absorption peak position and height are determined by the geometry of the metamaterial. Square-shaped metal/dielectric/metal stacks as absorber elements show spectral resonances at wavelengths between 4.8 and 7.0 microm in accordance with numerical simulations. Total peak absorptions of 0.8 are obtained. The metamaterial modified bolometers are suitable for multispectral thermal imaging systems in the mid-IR and terahertz regime.

Geometric Correction and Digital Elevation Extraction Using Multiple MTI Datasets

Digital Elevation Models (DEMs) are traditionally acquired from a stereo pair of aerial photographs sequentially captured by an airborne metric camera. Standard DEM extraction techniques can be naturally extended to satellite imagery, but the particular characteristics of satellite imaging can cause difficulties. The spacecraft ephemeris with respect to the ground site during image collects is the most important factor in the elevation extraction process. When the angle of separation between the stereo images is small, the extraction process typically produces measurements with low accuracy, while a large angle of separation can cause an excessive number of erroneous points in the DEM from occlusion of ground areas. The use of three or more images registered to the same ground area can potentially reduce these problems and improve the accuracy of the extracted DEM. The pointing capability of some sensors, such as the Multispectral Thermal Imager (MTI), allows for multiple collects of the same area from different perspectives. This functionality of MTI makes it a good candidate for the implementation of a DEM extraction algorithm using multiple images for improved accuracy. Evaluation of this capability and development of algorithms to geometrically model the MTI sensor and extract DEMs from multi-look MTI imagery are described in this paper. An RMS elevation error of 6.3-meters is achieved using 11 ground test points, while the MTI band has a 5-meter ground sample distance.

Data interpretation for spectral sensors with correlated bands

New classes of spectral sensors are emerging that have significant overlap in the band spectral response functions. While conventional sensors such as the Multispectral Thermal Images (MTI) or Landsat may have responses with a few percent overlap between adjacent bands, some of the emerging sensors can have more than 50% correlation among all spectral bands. The traditional geometrical models used to describe spectral data fail when such high levels of correlation exist. In this paper we present a generalized geometrical model that relies on functional analysis. We define a sensor space and a scene space that can be used to characterize the suitability of a sensor for a particular spectral sensing task. We demonstrate that classifiers based on first-order distance and angle metrics fail for sensors with highly correlated bands unless appropriate preprocessing is carried out. We further show that second-order statistical classifiers are largely immune to many of the problems introduced by the correlated band responses.

Satellite remote sensing of aerosols generated by the Island of Nauru

We use imagery from the Multispectral Thermal Imager (MTI) to search for aerosols generated by the Island of Nauru, an island located in the tropical western Pacific Ocean. Nauru frequently displays linear cloud trails for many kilometers downwind of the island, and this study was intended to investigate the presence of aerosols as an aid to understanding those features. The study had three components: (1) a search for specific aerosol plumes, (2) a comparison of downwind and upwind aerosol loading to look for asymmetries, and (3) application of matched filters to increase the visibility of aerosol plumes. The plume search resulted in the finding of three aerosol plumes, two of which are presented as imagery and also radiance profiles. The comparison of downwind and upwind reflectance spectra demonstrated that the radiance is slightly higher downwind of the island, and the residual spectra (downwind minus upwind) have a spectrum consistent with sea salt aerosol. Application of clutter‐matched filters to MTI imagery accentuated the upper (near source) sections of an aerosol plume when utilizing a single‐scatter albedo signature for coarse‐grained sea salt aerosol. We combine our observations with models from the literature to describe a simple mechanism by which Nauru aerosols are created by wave breaking and wind tearing of sea spray and then entrained by island‐influenced wind motions.

Comparison of near‐infrared and thermal infrared cloud phase detections

We compare the results of the cloud thermodynamic phase detections that use (1) the ratio of the near‐infrared and visible bands or (2) the brightness temperature difference of two thermal infrared bands. We find that the brightness temperature difference algorithm using the Moderate Resolution Imaging Spectroradiometer (MODIS) bands is generally consistent with the expectations based on the retrieval of the cloud top radiative temperature. On the other hand, the band ratio method, which uses near‐infrared and visible bands, assigns considerably more ice phase compared to the brightness temperature difference method and leads to discrepancies with the expectations based on the cloud top radiative temperature. When the cloud phase algorithm, developed originally for the Department of Energy Multispectral Thermal Imager research satellite, is applied to the MODIS imagery, the cloud phase assignments are close to the brightness temperature difference results and are in better agreement with the expectations based on the cloud top radiative temperature.

Extending surface temperature and emissivity retrieval to the mid-infrared (3–5 μm) using the Multispectral Thermal Imager (MTI)

Surface-temperature ( T) and emissivity ( ε) estimation from remotely sensed mid-infrared (MIR: 3–5 μm) data requires modifications to existing long-wave infrared (LWIR: 8–12 μm) T/ ε separation algorithms because of the significantly different characteristics of Planck's function between the MIR and LWIR wavelength regions and the strong effects of reflected solar irradiance in the MIR. A modified version of the normalized emissivity method (NEM), utilizing independently scaled maximum emissivities in each channel, was applied to thermal data acquired by the Multispectral Thermal Imager (MTI) with two MIR channels, J (3.81 μm) and K (4.97 μm), and three LWIR channels, L (8.26 μm), M (8.65 μm), and N (10.51 μm). Atmosphere-free simulations of T and ε retrieval over a wide variety of terrestrial surfaces yielded T values within ± 0.75 K of ‘true T’ in the range of 270–330 K and ε values within ± 0.011 of true ε in channels L, M and N, and ± 0.019 and ± 0.023 in channels K and J, respectively. The algorithm was tested successfully using MTI data over the Mauna Loa caldera in Hawaii. Unconstrained effects of shading and unresolved shadows in channel J day time data, and the strong atmospheric effects in channel K limit the application of the algorithm, in its present form, to night-time data.

In-flight validation of mid- and thermal infrared data from the Multispectral Thermal Imager (MTI) using an automated high-altitude validation site at Lake Tahoe CA/NV, USA

The Multispectral Thermal Imager (MTI) is a 15-band satellite-based imaging system. Two of the bands (J, K) are located in the mid-infrared (3-5 /spl mu/m) wavelength region: J, 3.5-4.1 /spl mu/m and K, 4.9-5.1 /spl mu/m, and three of the bands (L, M, N) are located in the thermal infrared (8-12 /spl mu/m) wavelength region: L, 8.0-8.4 /spl mu/m; M, 8.4-8.8 /spl mu/m; and N, 10.2-10.7 /spl mu/m. The absolute radiometric accuracy of the MTI data acquired in bands J-N was assessed over a period of approximately three years using data from the Lake Tahoe, CA/NV, automated validation site. Assessment involved using a radiative transfer model to propagate surface skin temperature measurements made at the time of the MTI overpass to predict the vicarious at-sensor radiance. The vicarious at-sensor radiance was convolved with the MTI system response functions to obtain the vicarious at-sensor MTI radiance in bands J-N. The vicarious radiances were then compared with the instrument measured radiances. In order to avoid any reflected solar contribution in the mid-infrared bands, only nighttime scenes were used in the analysis of bands J and K. Twelve cloud-free scenes were used in the analysis of the data from the mid-infrared bands (J, K), and 23 cloud-free scenes were used in the analysis of the thermal infrared bands (L, M, N). The scenes had skin temperatures ranging between 4.4 and 18.6/spl deg/C. The skin temperature was found to be, on average, 0.18/spl plusmn/0.36 degC cooler than the bulk temperature during the day and 0.65/spl plusmn/0.31 degC cooler than the bulk temperature at night. The smaller skin effect during the day was attributed to solar heating. The mean and standard deviation of the percent differences between the vicarious (predicted) at-sensor radiance convolved to the MTI bandpasses and the MTI measured radiances were -1.38/spl plusmn/2.32, -2.46/spl plusmn/1.96, -0.04/spl plusmn/0.78, -1.97/spl plusmn/0.62, -1.59/spl plusmn/0.55 for bands J-N, respectively. The results indicate that, with the exception of band L, the instrument measured radiances are warmer than expected.

Modeling the MTI electro-optic system sensitivity and resolution

We present an analysis methodology that offers efficient characterization of the Multispectral Thermal Imager (MTI) electro-optic system response to a wide range of user-specified system parameters and spectral scenarios. This methodology combines physics-based modeling of the MTI hardware with MTI prelaunch characterization data. The resulting models enable the user to generate application-specific sensitivity and resolution studies of the MTI image capture process, and aid in the development of calibration procedures and retrieval algorithms for MTI. In addition to quantifying the MTI response, the methodology developed in this paper is sufficiently general to permit the prototyping and evaluation of a variety of multispectral electro-optic systems. Finally, an example utilizing nominal orbital parameters and targeted MODTRAN scenarios that exercise the various spectral band functions is provided.

Multispectral thermal imager: mission and applications overview

The Multispectral Thermal Imager (MTI) satellite is a research and development project sponsored by the U.S. Department of Energy. The primary mission is to demonstrate advanced multispectral and thermal imaging from a satellite, including new technologies, data processing, and analysis techniques and validation by reference to ground truth. The MTI builds on the efforts of a number of earlier efforts, including Landsat, National Aeronautics and Space Administration remote sensing missions, and others, but the MTI incorporates a unique combination of attributes designed to advance the state of the art. The MTI satellite was launched on March 12, 2000 into a 580 km/spl times/610 km, sun-synchronous orbit with nominal 1 a.m. and 1 p.m. equatorial crossing times. The Air Force Space Test Program provided the Orbital Sciences Taurus launch. The satellite-based sensors obtain radiance data that are subsequently processed into measurements of atmospheric and surface properties such as column water vapor, atmospheric aerosol loading, surface temperatures, material composition, and others. This paper provides an overview of the MTI research objectives, design, operations, data products, and data processing and analysis. Several other papers provide greater detail on selected topics.

The effect of spatial resolution on satellite aerosol optical depth retrieval

We use data from the Multispectral Thermal Imager (MTI) to evaluate the effects of spatial resolution on the accuracy of aerosol optical depth (AOD) retrieval. Our results show that increasing the pixel size by itself has little effect on AOD retrieval accuracy at our chosen study site. However, increased pixel size does increase the error in AOD retrieval as a result of clouds. High-resolution sensors like MTI are able to avoid most clouds, but as the pixel size increases, subpixel clouds avoid the cloud mask, creep into selected pixels, and add a positive bias to the retrieved value of AOD. In accompanying work, we show that increasing pixel size has a small but noticeable impact on the normalized difference vegetation index (NDVI) and the 2.2-/spl mu/m reflectance, both used in the retrieval algorithm. We also examine the uniformity of the aerosol layer and show that the AOD varies by less than 0.02 in optical depth units over a 2.3/spl times/3.8 km/sup 2/ area. An analysis of the temporal variability of Aerosol Robotic Network-retrieved AOD shows a standard deviation of 0.02 on partly cloudy days and 0.004 on clear days.