Gotowa bibliografia na temat „Quasi-Hyperspectral Data”

Abstract. Measurements of hyperspectral canopy reflectance provide a detailed snapshot of information regarding canopy biochemistry, structure and physiology. In this study, we collected 5 years of repeated canopy hyperspectral reflectance measurements for a total of over 100 site visits within the flux footprints of two eddy covariance towers at a pasture and rice paddy in northern California. The vegetation at both sites exhibited dynamic phenology, with significant interannual variability in the timing of seasonal patterns that propagated into interannual variability in measured hyperspectral reflectance. We used partial least-squares regression (PLSR) modeling to leverage the information contained within the entire canopy reflectance spectra (400–900 nm) in order to investigate questions regarding the connection between measured hyperspectral reflectance and landscape-scale fluxes of net ecosystem exchange (NEE) and gross primary productivity (GPP) across multiple timescales, from instantaneous flux to monthly integrated flux. With the PLSR models developed from this large data set we achieved a high level of predictability for both NEE and GPP flux in these two ecosystems, where the R2 of prediction with an independent validation data set ranged from 0.24 to 0.69. The PLSR models achieved the highest skill at predicting the integrated GPP flux for the week prior to the hyperspectral canopy reflectance collection, whereas the NEE flux often achieved the same high predictive power at daily to monthly integrated flux timescales. The high level of predictability achieved by PLSR in this study demonstrated the potential for using repeated hyperspectral canopy reflectance measurements to help partition NEE into its component fluxes, GPP and ecosystem respiration, and for using quasi-continuous hyperspectral reflectance measurements to model regional carbon flux in future analyses.

Due to the extensive drilling performed every year in exploration campaigns for the discovery and evaluation of ore deposits, drill-core mapping is becoming an essential step. While valuable mineralogical information is extracted during core logging by on-site geologists, the process is time consuming and dependent on the observer and individual background. Hyperspectral short-wave infrared (SWIR) data is used in the mining industry as a tool to complement traditional logging techniques and to provide a rapid and non-invasive analytical method for mineralogical characterization. Additionally, Scanning Electron Microscopy-based image analyses using a Mineral Liberation Analyser (SEM-MLA) provide exhaustive high-resolution mineralogical maps, but can only be performed on small areas of the drill-cores. We propose to use machine learning algorithms to combine the two data types and upscale the quantitative SEM-MLA mineralogical data to drill-core scale. This way, quasi-quantitative maps over entire drill-core samples are obtained. Our upscaling approach increases result transparency and reproducibility by employing physical-based data acquisition (hyperspectral imaging) combined with mathematical models (machine learning). The procedure is tested on 5 drill-core samples with varying training data using random forests, support vector machines and neural network regression models. The obtained mineral abundance maps are further used for the extraction of mineralogical parameters such as mineral association.

We propose a new deep neural network termed TRQ3DNet which combines convolutional neural network (CNN) and transformer for hyperspectral image (HSI) denoising. The network consists of two branches. One is built by 3D quasi-recurrent blocks, including convolution and quasi-recurrent pooling operation. Specifically, the 3D convolution can extract the spatial correlation within a band, and spectral correlation between different bands, while the quasi-recurrent pooling operation is able to exploit global correlation along the spectrum. The other branch is composed of a series of Uformer blocks. The Uformer block uses window-based multi-head self-attention (W-MSA) mechanism and the locally enhanced feed-forward network (LeFF) to exploit the global and local spatial features. To fuse the features extracted by the two branches, we develop a bidirectional integration bridge (BI bridge) for better preserving the image feature information. Experimental results on synthetic and real HSI data show the superiority of our proposed network. For example, in the case of Gaussian noise with sigma 70, the PSNR value of our method significantly increases about 0.8 compared with other state-of-the-art methods.

High inflows of freshwater from the Mississippi and Atchafalaya rivers into the northern Gulf of Mexico during spring contribute to strong physical and biogeochemical gradients which, in turn, influence phytoplankton community composition across the river plume–ocean mixing zone. Spectral features representative of bio-optical signatures of phytoplankton size classes (PSCs) were retrieved from underway, shipboard hyperspectral measurements of above-water remote sensing reflectance using the quasi-analytical algorithm (QAA_v6) and validated against in situ pigment data and spectrophotometric analyses of phytoplankton absorption. The results shed new light on sub-km scale variability in PSCs associated with dynamic and spatially heterogeneous environmental processes in river-influenced oceanic waters. Our findings highlight the existence of localized regions of dominant picophytoplankton communities associated with river plume fronts in both the Mississippi and Atchafalaya rivers in an area of the coastal margin that is otherwise characteristically dominated by larger microphytoplankton. This study demonstrates the applicability of underway hyperspectral observations for providing insights about small-scale physical-biological dynamics in optically complex coastal waters. Fine-scale observations of phytoplankton communities in surface waters as shown here and future satellite retrievals of hyperspectral data will provide a novel means of exploring relationships between physical processes of river plume–ocean mixing and frontal dynamics on phytoplankton community composition.

The retrieval of sun-induced fluorescence (SIF) from hyperspectral radiance data grew to maturity with research activities around the FLuorescence EXplorer satellite mission FLEX, yet full-spectrum estimation methods such as the spectral fitting method (SFM) are computationally expensive. To bypass this computational load, this work aims to approximate the SFM-based SIF retrieval by means of statistical learning, i.e., emulation. While emulators emerged as fast surrogate models of simulators, the accuracy-speedup trade-offs are still to be analyzed when the emulation concept is applied to experimental data. We evaluated the possibility of approximating the SFM-like SIF output directly based on radiance data while minimizing the loss in precision as opposed to SFM-based SIF. To do so, we implemented a double principal component analysis (PCA) dimensionality reduction, i.e., in both input and output, to achieve emulation of multispectral SIF output based on hyperspectral radiance data. We then evaluated systematically: (1) multiple machine learning regression algorithms, (2) number of principal components, (3) number of training samples, and (4) quality of training samples. The best performing SIF emulator was then applied to a HyPlant flight line containing at sensor radiance information, and the results were compared to the SFM SIF map of the same flight line. The emulated SIF map was quasi-instantaneously generated, and a good agreement against the reference SFM map was obtained with a R2 of 0.88 and NRMSE of 3.77%. The SIF emulator was subsequently applied to 7 HyPlant flight lines to evaluate its robustness and portability, leading to a R2 between 0.68 and 0.95, and a NRMSE between 6.42% and 4.13%. Emulated SIF maps proved to be consistent while processing time was in the order of 3 min. In comparison, the original SFM needed approximately 78 min to complete the SIF processing. Our results suggest that emulation can be used to efficiently reduce computational loads of SIF retrieval methods.

The tropical dry forest (TDF) is one the most threatened ecosystems in South America, existing on a landscape with different levels of ecological succession. Among satellites dedicated to Earth observation and monitoring ecosystem succession, CHRIS/PROBA is the only satellite that presents quasi-simultaneous multi-angular pointing and hyperspectral imaging. These two characteristics permit the study of structural and compositional differences of TDFs with different levels of succession. In this paper, we use 2008 and 2014 CHRIS/PROBA images from a TDF in Minas Gerais, Brazil to study ecosystem succession after abandonment. Using a −55° angle of observation; several classifiers including spectral angle mapper (SAM), support vector machine (SVM), and decision trees (DT) were used to test how well they discriminate between different successional stages. Our findings suggest that the SAM is the most appropriate method to classify TDFs as a function of succession (accuracies ~80 for % for late stage, ~85% for the intermediate stage, ~70% for early stage, and ~50% for other classes). Although CHRIS/PROBA cannot be used for large-scale/long-term monitoring of tropical forests because of its experimental nature; our results support the potential of using multi-angle hyperspectral data to characterize the structure and composition of TDFs in the near future.

Narrow-band spectral indices are quite informative and important in various applications of remote sensing – to assess the condition of vegetation, soils, water bodies and other land surface formations. However, direct measurement of narrow-band spectral indices requires hyperspectral imaging. But most of modern multispectral aerospace imaging systems are wide-band. Accordingly, it is not possible to calculate the narrow-band index directly from wide-band remote sensing data. This paper discusses approaches to the narrow-band spectral indices restoration by wide-band remote sensing data using statistical models of interrelations of narrow- and wide-band indices itself, of source wide-band and narrow-band signals in close spectral bands, as well as of land surface reflectance quasi-continuous spectra translation from wide bands to narrow ones.The experimental accuracy estimation of narrow-band spectral indices restoration by wide-band multispectral satellite image is performed. Three most complicated narrow-band spectral indices, which covering a range of spectrum from visible to short-wave infrared, were considered, namely – the transformed chlorophyll absorption in reflectance index (TCARI), the optimized soil-adjusted vegetation index (OSAVI) and the normalized difference nitrogen index (NDNI). All three mentioned methods for narrow-band spectral indices restoration are analyzed. The worst result is demonstrated for regression-restored signals in spectral bands, and the best result is for the spectra translation method. Therefore, the method on the basis of spectra translation is recommended for practical implementation.

Hyperspectral electroluminescence and photoluminescence imaging of photovoltaic materials and devices produces three-dimensional spatially and spectrally resolved luminescence data, which can be calibrated to an absolute scale, enabling the extraction of high resolution maps of quantities, such as the local voltage (quasi-Fermi-level splitting). This extraction requires supplemental measurements of external quantum efficiency (EQE), but these do not have the same spatial resolution. Previously, assumptions have been made to overcome this limitation. In this work, we evaluate these assumptions for InGaAs solar cells with significant spatial variation in the luminescence spectrum shape due to small regions with elevated concentrations of radiative defects. Although appropriate for small variations in the spectral shape, we find that with more significant variation, these assumptions can result in nonphysical EQEs and too-low voltages. Combining multiple methods can help to alleviate this, or a minimum voltage map can be extracted, which will be similar to the actual voltage when EQE is high.

Abstract. Ammonia (NH3) emissions in the atmosphere have increased substantially over the past decades, largely because of intensive livestock production and use of fertilizers. As a short-lived species, NH3 is highly variable in the atmosphere and its concentration is generally small, except near local sources. While ground-based measurements are possible, they are challenging and sparse. Advanced infrared sounders in orbit have recently demonstrated their capability to measure NH3, offering a new tool to refine global and regional budgets. In this paper we describe an improved retrieval scheme of NH3 total columns from the measurements of the Infrared Atmospheric Sounding Interferometer (IASI). It exploits the hyperspectral character of this instrument by using an extended spectral range (800–1200 cm−1) where NH3 is optically active. This scheme consists of the calculation of a dimensionless spectral index from the IASI level1C radiances, which is subsequently converted to a total NH3 column using look-up tables built from forward radiative transfer model simulations. We show how to retrieve the NH3 total columns from IASI quasi-globally and twice daily above both land and sea without large computational resources and with an improved detection limit. The retrieval also includes error characterization of the retrieved columns. Five years of IASI measurements (1 November 2007 to 31 October 2012) have been processed to acquire the first global and multiple-year data set of NH3 total columns, which are evaluated and compared to similar products from other retrieval methods. Spatial distributions from the five years data set are provided and analyzed at global and regional scales. In particular, we show the ability of this method to identify smaller emission sources than those previously reported, as well as transport patterns over the ocean. The five-year time series is further examined in terms of seasonality and interannual variability (in particular as a function of fire activity) separately for the Northern and Southern Hemispheres.

In the recent years, remote sensing data or images have great potential for continuous spatial and temporal monitoring of Earth surface features. In case of optical remote sensing, hyperspectral (HS) data contains abundant spectral information and these information are advantageous for various applications. However, high-dimensional HS data handling is a very challenging task. Different techniques are proposed as a part of this thesis to handle the HS data in a computationally efficient manner and to achieve better performance for land cover classification and chlorophyll content prediction. Prior to start the HS data application, multispectral (MS) data are also analyzed in this thesis for crop classification. Multi-temporal MS data is used for crop classification. Landsat-8 operational land imager (OLI) sensor data are considered as MS data in this work. Surface reflectances and derived normalized difference indices (NDIs) of multi-temporal MS bands are combinedly used for the crop classification. Different dimensionality reduction techniques, viz. feature selection (FS) (e.g. random forest (RF) and partial informational correlation (PIC) measure-based), linear (e.g. principal component analysis (PCA) and independent component analysis) and nonlinear feature extraction (FE) (e.g. kernel PCA and Autoencoder), to be employed on the multi-temporal surface reflectances and NDIs datasets, are evaluated to detect the most favorable features. Subsequently, the detected features are used in a promising nonparametric classifier, support vector machine (SVM), for crop classification. It is found that all the evaluated FE techniques, employed on the multi-temporal datasets, resulted in better performance compared to FS-based approaches. PCA, being a simple and efficient FE algorithm, is well-suited in crop classification in terms of computational complexity and classification performances. Multi-temporal images are proved to be more advantageous compared to the single-date imagery for crop identification. HS data comprises of continuous spectral responses of hundreds of narrow spectral bands with very fine spectral resolution or bandwidth, which offer feature identification and classification with high accuracy. HS data are enriched with highly resourceful abundant spectral bands compared to only 5-10 spectral bands of MS data. However, analyzing and interpreting these ample amounts of data is a challenging task. Optimal spectral bands or features should be chosen or extracted to address the issue of redundancy and to capitalize on the absolute advantages of HS data. FS and FE are two broad categories of dimensionality reduction techniques. In this thesis, a FS and a FE-based computationally efficient dimensionality reduction technique is proposed for land cover classification. PIC-based HS band selection approach is proposed as a FS-based dimensionality reduction technique for classification of land cover types. PIC measure is more skillful compared to mutual information for estimation of non-parametric conditional dependency. In this proposed approach, HS narrow-bands are selected in an innovative way utilizing the PIC. Firstly, HS bands are divided into different spectral groups or segments using normalized mutual information (NMI) and then PIC is employed to each spectral group for optimal band selection. This approach is more efficient in terms of computational time and in generalizing the applicability of selected spectral bands. Further, these optimal spectral bands are used in the SVM and RF classifier for classification of land cover types and performance evaluation. The proposed FS-based dimensionality reduction approach is compared with different state-of-the-art techniques for land cover classification. The proposed methodology improved the classification performances compared to the existing techniques and the advancement in performances are proven to be statistically significant. In the recent years, deep learning-based FE techniques are very popular and also proven to be effective in extraction of apt features from the high-dimensional data. However, these techniques are computationally expensive. A computationally efficient FE-based dimensionality reduction approach, NMI-based segmented stacked auto-encoder (S-SAE), is proposed for extraction of spectral features from the HS data. These spectral features are consecutively utilized for creation of spatial features and later both spectral and spatial features are used in the classifier models (i.e. SVM and RF) for land cover classification. The proposed HS image classification approach reduces the complexity and computational time compared to the available techniques. A non-parametric dependency measure (i.e. NMI) based spectral segmentation is proposed instead of linear and parametric dependency measure to take care of the both linear and nonlinear inter-band dependencies for spectral segmentation of the HS bands. Then extended morphological profiles (EMPs) are created corresponding to segmented spectral features to assimilate the spatial information in the spectral-spatial classification approach. Two non-parametric classifiers, SVM with Gaussian kernel and RF are used for classification of the three most popularly used HS datasets. The experiments performed with the proposed methodology provide encouraging results compared to numerous existing approaches. HS data are proven to be more resourceful compared to MS data for object detection, classification and several other applications. However, absence of any space-borne HS sensor and high cost and limited obtainability of airborne sensors-based images limit the use of HS data. Transformation of readily available MS data into quasi-HS data can be a feasible solution for this issue. A deep learning-based regression algorithm, convolutional neural network regression (CNNR), is proposed as part of this thesis for MS (i.e. Landsat-7/8) to quasi-HS (i.e. quasi-Hyperion) data transformation. CNNR model introduces the advantages of nonlinear modelling and assimilation of spatial information in the regression-based modelling. The proposed CNNR model is compared with the pseudo-HS image transformation algorithm (PHITA), stepwise linear regression (SLR), and support vector regression (SVR) models by evaluating the quality of the quasi-Hyperion data. Several statistical metrics are calculated to compare each band’s reflectance values as well as spectral reflectance curve of each pixel of the quasi-Hyperion data with that of the original Hyperion data. The developed models and generated quasi-Hyperion data are also evaluated with application to crop classification. Analyzing the results of all the experiments, it is evident that CNNR model is more efficient compared to PHITA, SLR, and SVR in creating the quasi-Hyperion data and this transformed data are proven to be resourceful for crop classification application. The proposed CNNR model-based MS to quasi-HS data transformation approach can be used as a viable alternative for different applications in the absence of original HS images. HS data are investigated for estimation of chlorophyll content, which is one of the essential biochemical parameters to assess the growth process of the fruit trees. This study developed a model for estimation of canopy averaged chlorophyll content (CACC) of pear trees using the convolutional auto-encoder (CAE) features of HS data. This study also demonstrated the inspection of anomaly among the trees by employing multi-dimensional scaling (MDS) on the CAE features and detected the outlier trees, prior to fit nonlinear regression models. These outlier trees are excluded from further experiments which helped in improving the prediction performance of CACC. Gaussian process regression (GPR) and support vector regression (SVR) techniques are investigated as nonlinear regression models and used for prediction of CACC. The CAE features are proven to be providing better prediction of CACC, compared to the direct use of HS bands or vegetation indices as predictors. Training of the regression models, excluding the outlier trees, improved the CACC prediction performance. It is evident from the experiments that GPR can predict the CACC with better accuracy compared to SVR. In addition, the reliability of the tree canopy masks, which are utilized for averaging the features’ values for a particular tree, is also evaluated.