Hyperspectral Thermal Infrared Data Research Articles

New framework for land surface temperature and emissivity inversion based on hyperspectral thermal infrared data

ABSTRACT Hyperspectral thermal infrared (HTIR) data offers enormous potential for land surface temperature (LST) and emissivity (LSE) inversion. Methods to fully use a large amount of spectral information, while avoiding data redundancy, is constantly being explored. Therefore, a new framework for LST and LSE inversion using HTIR data is proposed in this study. First, the stepwise iteration method based on information content was used to select channels that were more sensitive to land surface information. Principal component analysis (PCA) and a two-step machine learning method were then used to retrieve the LST and LSE based on the selected channels. The findings demonstrate that the LST and spectra of the LSE can be obtained simultaneously based on the proposed framework. An LST inversion accuracy of 1.5K and an emissivity inversion accuracy of 0.017 can be achieved. However, an increase in the number of channels does not result in superior outcomes. Satisfactory inversion results can be obtained by selecting only 10–35 channels with the highest information content. In conclusion, this framework can be used to guide channel selection and land surface parameters inversion for HTIR sensors.

Temperature and Emissivity Retrieval From Hyperspectral Thermal Infrared Data Using Dictionary-Based Sparse Representation for Emissivity

The separation of land surface temperature (LST) and land surface emissivity (LSE) is an ill-posed problem in thermal infrared (TIR) remote sensing. By building a new observation matrix to compress the LSE unknows and a dictionary training method to reconstruct complete LSE spectra, a new dictionary-based sparse representation for emissivity (DSRE) method has been proposed to retrieve LST and LSE from the atmospherically corrected hyperspectral TIR data. The proposed method fully utilizes the sparsity of compressed sensing and the empirical knowledge of the trained emissivity dictionary. The sensitivity analysis shows that the modeling accuracies of the proposed method are 0.215 <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i> and 0.0060 for LST and LSE, respectively. Even with the instrument noise of 0.3 <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i> and the uncertainties in atmospheric transmittance, atmospheric upwelling, and downwelling radiance of 10 %, the retrieval accuracies are 0.811 <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i> for LST and 0.0241 for LSE, respectively. Then a field experiment was conducted to validate the proposed method, and a comparison was executed to three published methods, including ASTER temperature-emissivity separation (ASTERTES), linear spectral emissivity constraint TES (LSECTES), and iterative spectrally smooth TES (ISSTES). The accuracies of retrieved LST and spectral LSE are 1.41 <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i> / 0.009, 2.57 <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i> / 0.071, 1.59 <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i> / 0.038, and 2.00 <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i> / 0.077 for DSRE, ASTERTES, LSECTES, and ISSTES. In contrast to the three published methods, our proposed method is more accurate and effective than other published methods. Especially in the atmospheric absorption band, the proposed method has a strong anti-noise capability to the residuals of environmental downwelling radiance.

Simultaneous Estimation of Land Surface and Atmospheric Parameters From Thermal Hyperspectral Data Using a LSTM–CNN Combined Deep Neural Network

Thermal infrared (TIR) remote sensing observation signal is influenced by both atmospheric and land surface conditions that are difficult to separate with conventional multichannel TIR data. Because of the advantage of channel wealth, hyperspectral TIR data can simultaneously estimate the land surface and atmospheric parameters using neural network models or integrating them with physical models. However, the commonly used neural network models do not fully explore the correlation between different channels by treating the input data as discrete features. Thus, this study aims to develop a new deep neural network (DNN) by combining the long short-term memory (LSTM) network and convolutional neural network (CNN) for estimating land surface temperature (LST), emissivity, atmospheric transmittance, upward radiance, and downward radiance more accurately. By applying on the thermal airborne hyperspectral imager (TASI) simulation dataset covering global atmospheric conditions with 32 channels in 8.0-11.5 μm, the proposed model achieved results with the LST error of 0.95 K, the emissivity error of less than 0.012 for each channel, and the accuracy of three atmospheric parameters has also been improved compared with the current neural network models. Our model has been applied to a real TASI image, and its validity was further proved by the ground measurement validation data. Therefore, it can provide more reliable initial values for physical optimization models.

Lithology Classification Using TASI Thermal Infrared Hyperspectral Data with Convolutional Neural Networks

In recent decades, lithological mapping techniques using hyperspectral remotely sensed imagery have developed rapidly. The processing chains using visible-near infrared (VNIR) and shortwave infrared (SWIR) hyperspectral data are proven to be available in practice. The thermal infrared (TIR) portion of the electromagnetic spectrum has considerable potential for mineral and lithology mapping. In particular, the abovementioned rocks at wavelengths of 8–12 μm were found to be discriminative, which can be seen as a characteristic to apply to lithology classification. Moreover, it was found that most of the lithology mapping and classification for hyperspectral thermal infrared data are still carried out by traditional spectral matching methods, which are not very reliable due to the complex diversity of geological lithology. In recent years, deep learning has made great achievements in hyperspectral imagery classification feature extraction. It usually captures abstract features through a multilayer network, especially convolutional neural networks (CNNs), which have received more attention due to their unique advantages. Hence, in this paper, lithology classification with CNNs was tested on thermal infrared hyperspectral data using a Thermal Airborne Spectrographic Imager (TASI) at three small sites in Liuyuan, Gansu Province, China. Three different CNN algorithms, including one-dimensional CNN (1-D CNN), two-dimensional CNN (2-D CNN) and three-dimensional CNN (3-D CNN), were implemented and compared to the six relevant state-of-the-art methods. At the three sites, the maximum overall accuracy (OA) based on CNNs was 94.70%, 96.47% and 98.56%, representing improvements of 22.58%, 25.93% and 16.88% over the worst OA. Meanwhile, the average accuracy of all classes (AA) and kappa coefficient (kappa) value were consistent with the OA, which confirmed that the focal method effectively improved accuracy and outperformed other methods.

Retrieval of Land Surface Temperature, Emissivity, and Atmospheric Parameters From Hyperspectral Thermal Infrared Image Using a Feature-Band Linear-Format Hybrid Algorithm

Thermal infrared remote sensing can acquire large-scale land surface thermal radiance effectively. However, the observed data are affected by surface and atmospheric conditions. Traditional methods require some prior knowledge, such as emissivity in split-window algorithm and atmospheric correction in temperature–emissivity separation algorithm. This information is difficult to obtain directly and accurately. Hyperspectral thermal infrared data provide the possibility for simultaneous retrieval of atmospheric parameters, land surface temperature (LST), and emissivity because of their abundant band information. This study proposed a feature-band linear-format hybrid (FebLihy) algorithm by combining a deep neural network (DNN) model and a physical model with thermal airborne hyperspectral imager (TASI) data. The proposed algorithm was divided into three steps. First, the radiative transfer equation was converted into a linear form, and seven feature bands were chosen to reduce the unknowns. Second, the initial values of atmospheric and land surface parameters were estimated with the DNN model. Finally, least-squares optimization was used in the physical model to retrieve the final results. Results of the simulation data showed that the root-mean-square error (RMSE) of LST was 0.86 K, the RMSE of emissivity was less than 0.015, and the accuracy of atmospheric parameters was improved effectively by the physical model. The FebLihy algorithm was applied in a real TASI image in Fuyun County and verified with CE312 ground measurement data. Accurate results were achieved. The FebLihy algorithm will be optimized in terms of model and data in the future study.

Noise-sensitivity Analysis and Improvement of Automatic Retrieval of Temperature and Emissivity Using Spectral Smoothness

There are numerous algorithms that can be used to retrieve land surface temperature (LST) and land surface emissivity (LSE) from hyperspectral thermal infrared (HTIR) data. The algorithms are sensitive to a number of factors, where noise is difficult to handle due to its unpredictability. Although there is a lot of research regarding the influence of noise on retrieval errors, few studies have focused on the mechanism. In this study, we selected the automatic retrieval of temperature and emissivity using spectral smoothness (ARTEMISS) algorithm—the representative of the iterative spectral smoothness temperature-emissivity separation algorithm family—as the research object and proposed an improved algorithm. First, we analyzed the influence mechanism of noise on the retrieval errors of ARTEMISS in theory. Second, we carried out a simulation and inversion experiment and analyzed the relationship between instrument spectral resolution, noise level, the ARTEMISS parameter setting and the retrieval errors separately. Last, we proposed an improved method (resolution-degrade-based spectral smoothness algorithm, RDSS) based on the mechanism and law of the influence of noise on retrieval errors and provided corresponding suggestions on instrument design. The results show that RDSS improves the accuracy of temperature inversion and is more effective for thermal infrared data with a high noise level and high spectral resolution, which can reduce the LST inversion error by up to 0.75 K and the LSE median absolute deviation (MAD) by 31%. In the presence of noise in HTIR data, the RDSS algorithm performs better than the ARTEMISS algorithm in terms of temperature-emissivity separation.

Temperature and Emissivity Inversion Accuracy of Spectral Parameter Changes and Noise of Hyperspectral Thermal Infrared Imaging Spectrometers

The emergence of hyperspectral thermal infrared imaging spectrometers makes it possible to retrieve both the land surface temperature (LST) and the land surface emissivity (LSE) simultaneously. However, few articles focus on the problem of how the instrument’s spectral parameters and instrument noise level affect the LST and LSE inversion errors. In terms of instrument development, this article simulated three groups of hyperspectral thermal infrared data with three common spectral parameters and each group of data includes tens of millions of simulated radiances of 1525 emissivity curves with 17 center wavelength shift ratios, 6 full width at half maximum (FWHM) change ratios and 6 noise equivalent differential temperatures (NEDTs) under 15 atmospheric conditions with 6 object temperatures, inverted them by two temperature and emissivity separation methods (ISSTES and ARTEMISS), and analyzed quantitatively the effects of the spectral parameters change and noise of an instrument on the LST and LSE inversion errors. The results show that: (1) center wavelength shifts and noise affect the inversion errors strongly, while FWHM changes affect them weakly; (2) the LST and LSE inversion errors increase with the center wavelength shift ratio in a quadratic function and increase with FWHM change ratio slowly and linearly for both the inversion methods, however they increase with NEDT in an S-curve for ISSTES while they increase with NEDT slightly and linearly for ARTEMISS. During the design and development of a hyperspectral thermal infrared instrument, it is highly recommended to keep the potential center wavelength shift within 1 band and keep NEDT within 0.1K (corresponding LST error < 1K and LSE error < 0.015) for normal applications and within 0.03K (corresponding LST error < 0.5K and LSE error < 0.01) for better application effect and level.

Mineral Identification and Mapping by Synthesis of Hyperspectral VNIR/SWIR and Multispectral TIR Remotely Sensed Data With Different Classifiers

Hyperspectral data, which have fine continuous spectrum, have been recognized to be more suitable for the detailed identification and classification of land surface, especially for minerals. The combination of the hyperspectral visible/near-infrared (VNIR) and shortwave infrared (SWIR) data with the hyperspectral thermal infrared (TIR) data is proven to be an effective way. However, how those effects are and what are the effects of introduction of multispectral TIR data on the minerals identification and classification are not well studied. To fully evaluate those effects, this article tries to use both simulated data and real data to testify the practicability of introduction of multispectral TIR data for the accuracies of mineral identification and classification. Four classifiers, i.e., spectral angle mapping, spectral feature fitting, orthogonal subspace projection, and adaptive coherence/cosine estimator, are selected in the experiment. Compared with the results using hyperspectral data alone, the introducing of multispectral TIR data in identification and classification has improved accuracies for both the simulated and real data. The overall accuracies are improved about 4%–13% for the simulated data and about 1%–5% for the real data by using different classifiers. Those improvements prove that the spectral diagnosed characteristics in TIR region even for multispectral data help identify and classify minerals. Although the improvements for real data are not well obvious due to the low spatial resolution, the multispectral TIR data are still effective supplements for hyperspectral VNIR and SWIR data in mineral identification and classification.

An Improved Linear Spectral Emissivity Constraint Method for Temperature and Emissivity Separation Using Hyperspectral Thermal Infrared Data.

The linear spectral emissivity constraint (LSEC) method has been proposed to separate temperature and emissivity in hyperspectral thermal infrared data with an assumption that land surface emissivity (LSE) can be described by an equal interval piecewise linear function. This paper combines a pre-estimate shape method with the LSEC method to provide an initial-shape estimation of LSE which will create a new piecewise scheme for land surface temperature (LST) and LSE separation. This new scheme is designated as the pre-estimate shape (PES)-LSEC method. Comparisons with the LSEC method using simulated data sets show that the PES-LSEC method has better performance in terms of accuracy for both LSE and LST. With an at-ground error of 0.5 K, the root-mean-square errors (RMSEs) of LST and LSE are 0.07 K and 0.0045, respectively, and with the scale factor of moisture profile 0.8 and 1.2, the RMSEs of LST are 1.11 K and 1.14 K, respectively. The RMSEs of LSE in each channel are mostly below 0.02 and 0.04, respectively, which are better than for the LSEC method. In situ experimental data are adopted to validate our method: The results show that RMSE of LST is 0.9 K and the mean value of LSE accuracy is 0.01. The PES-LSEC method with fewer segments achieves better accuracy than that of LSEC and preserves most of the crest and trough information of emissivity.

Retrieving Atmospheric and Land Surface Parameters From At-Sensor Thermal Infrared Hyperspectral Data With Artificial Neural Network

The radiances observed by satellites are influenced by both land surface and atmospheric parameters, and it is difficult to retrieve these parameters simultaneously from multispectral measurements with high accuracies. Even though several methods have been proposed, they focus on the retrieval of land surface or atmospheric parameters separately. Generally, these atmospheric parameters are atmospheric water vapor and temperature profiles. Thus, this study aims to establish a back propagation (BP) artificial neural network (ANN) to retrieve land surface emissivity (LSE), land surface temperature (LST), atmospheric transmittance, upward radiance, and downward radiance simultaneously from the hyperspectral thermal infrared (TIR) data, suitable for various air mass types and surface conditions. The principle component analysis technique is first used to compress and remove noise from the data. The evaluation of the ANN using the simulated data without noise indicated that the root mean square error (RMSE) of LST is approximately 0.643 K; the RMSEs of emissivity, transmittance, upward, and downward radiance are approximately 0.0046, 0.005, 0.72, and 2.95 K, respectively. When applied on the simulated data containing noise, the errors of LST, LSE, transmittance, upward, and downward radiance are 1.26, 0.01, 0.01, 1.54, and 4.57 K, respectively. When applied on the real atmospheric infrared sounder data, the retrieved accuracies become worse because of various unstudied reasons. However, the results show that the proposed ANN is promising in retrieving the land surface and atmospheric parameters simultaneously. Because of its simplicity, the proposed ANN can be used to produce preliminary results employed as the first estimates for physics-based retrieval methods.

Mapping the wavelength position of mineral features in hyperspectral thermal infrared data

The Wavelength Mapper is an algorithm that searches for the deepest absorption feature in each pixel of a hyperspectral image. On a per pixel basis, it extracts the wavelength position, which serves as a proxy of the mineralogy and the feature depth as a proxy for the relative abundance. This algorithm has been used with near and shortwave infrared data, but has not yet been tested on hyperspectral thermal infrared images. It is unclear what results are expected when the Wavelength Mapper algorithm is applied to hyperspectral thermal infrared data since reststrahlen features characteristically overlap in emissivity spectra. In this paper, the Wavelength Mapper is tested on a multi-flightline airborne hyperspectral TIR dataset acquired over the Yerington Batholith, Nevada. Observations were made in the 8.05–11.65 μm wavelength range to include thermal spectral features of major rock-forming minerals, and a new color ramp is created to separate quartz-rich rocks from plagioclase-rich rocks. Our results indicate that the Wavelength Mapper creates coherent spatial patterns across flightlines. The results displayed represent different types of igneous and sedimentary rocks, as well as the products of hydrothermal alteration via different colors, mainly based on the relative abundance of quartz, feldspar and garnet, as well as mica and epidote. Comparison with published maps indicate that the Wavelength Mapper represents for each pixel a parameter value that can be linked to the spectrally dominate rock-forming mineral of that area, as mapped with traditional fieldwork methods. In conclusion, the Wavelength Mapper can be applied to airborne hyperspectral TIR data to achieve a simple, repeatable, per-pixel overview map of the dominating rock-forming mineral occurrences.

Temperature and emissivity retrieval from low-emissivity materials using hyperspectral thermal infrared data

ABSTRACTLand surface temperature and emissivity separation (TES) is a key problem in thermal infrared remote sensing. Many TES algorithms have been proposed and validated on natural surface materials such as water or vegetation. However, when those algorithms are applied to low-emissivity materials such as some metals, the retrieval accuracy needs significant improvement. Here, we propose a new temperature and emissivity retrieval algorithm for hyperspectral thermal infrared data based on bias characteristic of atmospheric downward radiance. This approach is aimed at weakening the large influence of errors in atmospheric downward radiance on the TES of low-emissivity materials due to stronger coupling between the surface and atmosphere. Compared with traditional approaches such as the iterative spectrally smooth temperature and emissivity separation (ISSTES) method, the proposed method improves retrieval accuracy for low-emissivity materials by about 0.3, 0.4, and 1 K for three low-emissivity test materials (galvanized steel metal, metallic silver paint, and gold paint sandpaper). Thus, the proposed algorithm can weaken the influence of atmospheric downward radiance error. In addition, the computation efficiency is higher than ISSTES because fewer groups of channels are involved in the calculation.

Evaluation and comparison of hyperspectral temperature and emissivity separation methods influenced by sensor spectral properties

ABSTRACTIn this article, great efforts are focused on the temperature and emissivity separation (TES) from hyperspectral thermal infrared data. Instead of proposing a new TES method, several published TES methods, including iterative spectrally smooth temperature–emissivity separation method (ISSTES), automatic retrieval of temperature and emissivity using spectral smoothness method (ARTEMISS), spectral smoothness method (SpSm), downwelling radiance residual index method (DRRI), and linear spectral emissivity constraint method (LSEC), are reviewed. The performances of these TES methods are analysed and compared under different instrument characteristics, including spectral shifting and full-width half-maximum broadening, with the simulated data. The results show that LSEC has the most robust and accurate performance. DRRI also has a good performance, in which a channel selection procedure is required before the use of this method. ISSTES, ARTEMISS, and SpSm show more sensitivities.

A multi-scale wavelet-based temperature and emissivity separation algorithm for hyperspectral thermal infrared data

ABSTRACTLand surface temperature and emissivity are essential variables in numerous environmental studies. This article proposes a multi-scale wavelet-based temperature and emissivity separation (MSWTES) algorithm. MSWTES is based on the fact that the high frequencies of ground-leaving radiance and derived emissivity spectra using inaccurate temperature are both closely correlated with the atmospheric downward radiance spectrum. First, surface emissivity can be decomposed by multi-scale wavelet into an optimal level that can be derived from correlation between reconstructed high frequency of ground-leaving radiance and atmospheric downward radiance. Then the ratio of high-frequency energy to low-frequency energy of surface emissivity spectrum is used to measure the degree of atmospheric downward radiance residue in the calculated emissivity spectrum as well as the disparity between the initial surface temperature and the true value. Finally, we can derive the optimal estimate of surface temperature and calculate the surface emissivity spectrum accordingly with this criterion. The MSWTES is first tested by simulation data. When a noise-equivalent spectral error of 2.5 × 10–9 W cm−2 sr−1 cm is considered, the average temperature bias is 0.027 K and the root mean square error (RMSE) of emissivity is less than 0.003, except at the low and high ends of the 750–1250 cm−1 spectral region. Then, the MSWTES is applied to field measurements. As a whole, the MSWTES achieves an RMSE of 0.01 for emissivity retrieval under most conditions, but its accuracy degrades when sample emissivity is extremely low. Meanwhile, the MSWTES is compared to the iterative spectrally smooth temperature and emissivity separation (ISSTES) algorithm. The performance of the MSWTES is better than that of the ISSTES, which demonstrates the good performance of the MSWTES.

Land Surface Temperature Estimate From Chinese Gaofen-5 Satellite Data Using Split-Window Algorithm

The Gaofen-5 (GF-5) satellite, the only satellite that provides the thermal infrared (TIR) sensor in the national high-resolution earth observation project of China, will observe earth surface at a spatial resolution of 40 m in four TIR channels. This paper aims at developing a new nonlinear, four-channel split-window (SW) algorithm to retrieve land surface temperature (LST) from GF-5 image. In the SW algorithm, its coefficients were obtained based on several subranges of atmospheric column water vapors (CWV) under various land surface conditions, in order to remove the atmospheric effect and improve the retrieval accuracy. Results showed that the new algorithm can obtain LST with root-mean-square errors of less than 1 K. Compared with previous two- and three-channel SW algorithms, the four-channel SW algorithm obtained better results in estimating LST, especially under moist atmospheres. Methods of estimating CWV and pixel emissivity were also conducted. The sensitive analysis of LST retrieval to instrument noise and uncertainty of pixel emissivity and water vapor demonstrated the good performance of the proposed algorithm. At last, the new SW algorithm was validated using ground-measured data at six sites, and some simulated images from airborne hyperspectral TIR data.

Evaluation of thermal infrared hyperspectral imagery for the detection of onshore methane plumes: Significance for hydrocarbon exploration and monitoring

Methane (CH4) is the main constituent of natural gas. Fugitive CH4 emissions partially stem from geological reservoirs (seepages) and leaks in pipelines and petroleum production plants. Airborne hyperspectral sensors with enough spectral and spatial resolution and high signal-to-noise ratio can potentially detect these emissions. Here, a field experiment performed with controlled release CH4 sources was conducted in the Rocky Mountain Oilfield Testing Center (RMOTC), Casper, WY (USA). These sources were configured to deliver diverse emission types (surface and subsurface) and rates (20–1450 scf/hr), simulating natural (seepages) and anthropogenic (pipeline) CH4 leaks. The Aerospace Corporation’s SEBASS (Spatially-Enhanced Broadband Array Spectrograph System) sensor acquired hyperspectral thermal infrared data over the experimental site with 128 bands spanning the 7.6μm–13.5μm range. The data was acquired with a spatial resolution of 0.5m at 1500ft and 0.84m at 2500ft above ground level. Radiance images were pre-processed with an adaptation of the In-Scene Atmospheric Compensation algorithm and converted to emissivity through the Emissivity Normalization algorithm. The data was processed with a Matched Filter. Results allowed the separation between endmembers related to the spectral signature of CH4 from the background. Pixels containing CH4 signatures (absorption bands at 7.69μm and 7.88μm) were highlighted and the gas plumes mapped with high definition in the imagery. The dispersion of the mapped plumes is consistent with the wind direction measured independently during the experiment. Variations in the dimension of mapped gas plumes were proportional to the emission rate of each CH4 source. Spectral analysis of the signatures within the plumes shows that CH4 spectral absorption features are sharper and deeper in pixels located near the emitting source, revealing regions with higher gas density and assisting in locating CH4 sources in the field accurately. These results indicate that thermal infrared hyperspectral imaging can support the oil industry profusely, by revealing new petroleum plays through direct detection of gaseous hydrocarbon seepages, serving as tools to monitor leaks along pipelines and oil processing plants, while simultaneously refining estimates of CH4 emissions.

Land Surface Temperature and Emissivity Retrieval from Field-Measured Hyperspectral Thermal Infrared Data Using Wavelet Transform

Currently, the main difficulty in separating the land surface temperature (LST) and land surface emissivity (LSE) from field-measured hyperspectral Thermal Infrared (TIR) data lies in solving the radiative transfer equation (RTE). Based on the theory of wavelet transform (WT), this paper proposes a method for accurately and effectively separating LSTs and LSEs from field-measured hyperspectral TIR data. We show that the number of unknowns in the RTE can be reduced by decomposing and reconstructing the LSE spectrum, thus making the RTE solvable. The final results show that the errors introduced by WT are negligible. In addition, the proposed method usually achieves a greater accuracy in a wet-warm atmosphere than that in a dry-cold atmosphere. For the results under instrument noise conditions (NE∆T = 0.2 K), the overall accuracy of the LST is approximately 0.1–0.3 K, while the Root Mean Square Error (RMSE) of the LSEs is less than 0.01. In contrast to the effects of instrument noise, our method is quite insensitive to noises from atmospheric downwelling radiance, and all the RMSEs of our method are approximately zero for both the LSTs and the LSEs. When we used field-measured data to better evaluate our method’s performance, the results showed that the RMSEs of the LSTs and LSEs were approximately 1.1 K and 0.01, respectively. The results from both simulated data and field-measured data demonstrate that our method is promising for decreasing the number of unknowns in the RTE. Furthermore, the proposed method overcomes some known limitations of current algorithms, such as singular values and the loss of continuity in the spectrum of the retrieved LSEs.

A novel method for urban land cover mapping based on new vegetation indices and texture-spectral information from fused visible and hyperspectral thermal infrared airborne data

ABSTRACTUpdate information on urban regions has been substantial for management communities. In this research, a novel method was developed for Urban Land Cover Mapping (ULCM). Textural-spectral features obtained from Hyperspectral Thermal Infrared (HTIR) data were fused with spatial-spectral features of the visible image for ULCM. The proposed method consists of three hierarchical steps. First, trees and vegetation classes were classified based on spatial-spectral features extracted from the visible image. Also, two new vegetation indices were introduced. By studying spectral signatures of trees and vegetation classes on the HTIR data, it was shown that trees and vegetation classes could be discriminated by visible data. Second, textural-spectral features of the HTIR data were fused with visible image features to extract bare soil, (gray; concrete; red) roof buildings and roads classes. Using HTIR textural features increased the overall accuracy and Kappa coefficient values about 6% and 8% correspondingly. Third, the results of the first and second steps were overlaid and post processing has been done. The obtained results for overall accuracy and Kappa coefficient values were 94.96% and 0.928 respectively. The comparison of the achieved results with the results of the contest announced by IEEE has shown the efficiency of the proposed method.

Evaluation of Temperature and Emissivity Retrieval using Spectral Smoothness Method for Low-Emissivity Materials

Land surface temperature and emissivity separation (TES) is a key problem in thermal infrared (TIR) remote sensing. Along with the development of civil applications, increasing numbers of man-made low-emissivity materials can be found around our living environment. In addition, the characteristics and variation in properties of those materials should also be concerned. However, there are still few TES methods for low-emissivity materials reported in the literature. This paper addresses the performance of the automatic retrieval of temperature and emissivity using spectral smoothness (ARTEMISS) method proposed by Borel (2008) for the retrieval of temperature and emissivity from hyperspectral TIR data for low-emissivity materials. The results show that those modeling errors are less than 0.11 K for temperature and 0.3% for emissivity as shown in the ARTEMISS algorithm if atmospheric parameters and the mean emissivity of material spectra are known. A sensitivity analysis has been performed, and the results show that the retrieval accuracy will be degraded with the increase of instrument noises, the errors of the atmospheric parameters, and the coarser spectral resolution. ARTEMISS can give a reasonable estimation of the temperature and emissivity for high- and low-emissivity materials; however, the performance of the algorithm is more seriously influenced by the atmospheric compensation than by the instrument noises. Our results show that the errors of temperature and emissivity become approximately three times than that when the instrument spectral properties are $1{\text{ cm}}^{-1}$ of sampling interval and $2{\text{ cm}}^{-1}$ of FWHM, and $4{\text{ cm}}^{-1}$ of sampling interval and $8{\text{ cm}}^{-1}$ of FWHM, respectively.

DETECTION AND DISCRIMINATION OF THE THICK OIL PATCHES ON THE SEA SURFACE

Detection of natural or accidental oil slick at sea surface is important both for exploration purposes and for environment protection. Radar imagery, either satellite or airborne is the prime tool to detect those slicks. Radar is widely used by the national agencies to monitor their maritime areas for accidental pollutions or boat discharges. Radar images can detect oil slick even in the presence of clouds. However the sea surface back scattered energy is rather insensitive to oil thickness. On the contrary several studies tend to prove that optical data may be used to estimate the oil thickness. These data may be in the form of hyperspectral data or thermal infrared data. The objective of this study is to show that SWIR satellite data which are more widely available than hyperspectral data, better resolved than thermal data and available at a very limited cost, can be used to detect and qualitatively assess the thickness of oil slicks. This is important to assess volumes of naturally release oil in the oceans and in case of a crisis to send intervention teams where oil is thickest.