Due to advent of sensor technology, hyperspectral imaging has become an emerging technology in remote sensing. Many problems, which cannot be resolved by multispectral imaging, can now be solved by hyperspectral imaging. The aim of this Special Issue “Hyperspectral Imaging and Applications” is to publish new ideas and technologies to facilitate the utility of hyperspectral imaging in data exploitation and to further explore its potential in different applications. This Special Issue has accepted and published 25 papers in various areas, which can be organized into 7 categories, Data Unmixing, Spectral variability, Target Detection, Hyperspectral Image Classification, Band Selection, Data Fusion, Applications. Full article

Real-time target detection for hyperspectral images (HSI) has received considerable interest in recent years. However, owing to enormous data volume provided by HSI, detection algorithms are generally computationally complex, thus developing rapid processing techniques for target detection has encountered several challenging issues. It seems that using a deep pipelined structure can improve the detection speed, and implementing on field programmable gate arrays (FPGAs) can also achieve concurrent operations rather than run streams of sequential instruction. This paper presents a deep pipelined background statistics (DPBS) approach to optimizing and implementing a well-known subpixel target detection algorithm, called constrained energy minimization (CEM) on FPGA by using high-level synthesis (HLS). This approach offers significant benefits in terms of increasing data throughput and improving design efficiency. To overcome a drawback of HLS on implementing a task-level pipelined circuit that includes a feedback data path, a script based circuit design method is further developed to make connections between some of the modules created by HLS. Experimental results show that the proposed method can detect targets on a real-hyperspectral data set (HyMap Data) only in 0.15 s without compromising detection accuracy. Full article

Semi-supervised classification methods result in higher performance for hyperspectral images, because they can utilize the relationship between unlabeled samples and labeled samples to obtain pseudo-labeled samples. However, how generating an effective training sample set is a major challenge for semi-supervised methods, In this paper, we propose a novel semi-supervised classification method based on extended label propagation (ELP) and a rolling guidance filter (RGF) called ELP-RGF, in which ELP is a new two-step process to make full use of unlabeled samples. The first step is to implement the graph-based label propagation algorithm to propagate the label information from labeled samples to the neighboring unlabeled samples. This is then followed by the second step, which uses superpixel propagation to assign the same labels to all pixels within the superpixels that are generated by the image segmentation method, so that some labels wrongly labeled by the above step can be modified. As a result, so obtained pseudo-labeled samples could be used to improve the performance of the classifier. Subsequently, an effective feature extraction method, i.e., RGF is further used to remove the noise and the small texture structures to optimize the features of the initial hyperspectral image. Finally, these produced initial labeled samples and high-confidence pseudo-labeled samples are used as a training set for support vector machine (SVM). The experimental results show that the proposed method can produce better classification performance for three widely-used real hyperspectral datasets, particularly when the number of training samples is relatively small. Full article

Component substitution (CS) and multiresolution analysis (MRA) based methods have been adopted in hyperspectral pansharpening. The major contribution of this paper is a novel CS-MRA hybrid framework based on intrinsic image decomposition and weighted least squares filter. First, the panchromatic (P) image is sharpened by the Gaussian-Laplacian enhancement algorithm to enhance the spatial details, and the weighted least squares (WLS) filter is performed on the enhanced P image to extract the high-frequency information of the P image. Then, the MTF-based deblurring method is applied to the interpolated hyperspectral (HS) image, and the intrinsic image decomposition (IID) is adopted to decompose the deblurred interpolated HS image into the illumination and reflectance components. Finally, the detail map is generated by making a proper compromise between the high-frequency information of the P image and the spatial information preserved in the illumination component of the HS image. The detail map is further refined by the information ratio of different bands of the HS image and injected into the deblurred interpolated HS image. Experimental results indicate that the proposed method achieves better fusion results than several state-of-the-art hyperspectral pansharpening methods. This demonstrates that a combination of an IID technique and a WLS filter is an effective way for hyperspectral pansharpening. Full article

In this paper, we introduce a novel classification framework for hyperspectral images (HSIs) by jointly employing spectral, spatial, and hierarchical structure information. In this framework, the three types of information are integrated into the SVM classifier in a way of multiple kernels. Specifically, the spectral kernel is constructed through each pixel’s vector value in the original HSI, and the spatial kernel is modeled by using the extended morphological profile method due to its simplicity and effectiveness. To accurately characterize hierarchical structure features, the techniques of Fish-Markov selector (FMS), marker-based hierarchical segmentation (MHSEG) and algebraic multigrid (AMG) are combined. First, the FMS algorithm is used on the original HSI for feature selection to produce its spectral subset. Then, the multigrid structure of this subset is constructed using the AMG method. Subsequently, the MHSEG algorithm is exploited to obtain a hierarchy consist of a series of segmentation maps. Finally, the hierarchical structure information is represented by using these segmentation maps. The main contributions of this work is to present an effective composite kernel for HSI classification by utilizing spatial structure information in multiple scales. Experiments were conducted on two hyperspectral remote sensing images to validate that the proposed framework can achieve better classification results than several popular kernel-based classification methods in terms of both qualitative and quantitative analysis. Specifically, the proposed classification framework can achieve 13.46–15.61% in average higher than the standard SVM classifier under different training sets in the terms of overall accuracy. Full article

In this paper, a new sliding window-based joint sparse representation (SWJSR) anomaly detector for hyperspectral data is proposed. The main contribution of this paper is to improve the judgments about the probability of anomaly presence in signals using the integration of information gathered during transition of sliding window for each pixel. In this method, each pixel experiences different spatial positions with respect to the spatial neighbors through the transition of this sliding window. In each position, an optimized local background dictionary is formed using a K-Singular Value Decomposition (K-SVD) algorithm and the recovery error of sparse estimation for each pixel is calculated using a simultaneous orthogonal matching pursuit algorithm (SOMP). Thus, the votes of each signal in terms of the anomaly presence in each spatial neighborhood are calculated and the variance of these recovery errors is considered as the detection criterion. The experimental results of the proposed SWJSR method on both synthetic and real datasets proved its higher performance compared to the Global RX (GRX), Local RX (LRX), Collaborative Representation Detector (CRD), Background Joint Sparse Representation (BJSR), Causal RX Detector (CR-RXD, CK-RXD), and Sliding Local RX(SLRX) detectors with an average efficiency improvement of about 7.5%, 14.25%, 8.2%, 8.25%, 6.45%, 6.5%, and 3.6%, respectively, in comparison to the mentioned algorithms. Full article

Hyperspectral sensors are able to provide information that is useful for many different applications. However, the huge amounts of data collected by these sensors are not exempt of drawbacks, especially in remote sensing environments where the hyperspectral images are collected on-board satellites and need to be transferred to the earth’s surface. In this situation, an efficient compression of the hyperspectral images is mandatory in order to save bandwidth and storage space. Lossless compression algorithms have been traditionally preferred, in order to preserve all the information present in the hyperspectral cube for scientific purposes, despite their limited compression ratio. Nevertheless, the increment in the data-rate of the new-generation sensors is making more critical the necessity of obtaining higher compression ratios, making it necessary to use lossy compression techniques. A new transform-based lossy compression algorithm, namely Lossy Compression Algorithm for Hyperspectral Image Systems (HyperLCA), is proposed in this manuscript. This compressor has been developed for achieving high compression ratios with a good compression performance at a reasonable computational burden. An extensive amount of experiments have been performed in order to evaluate the goodness of the proposed HyperLCA compressor using different calibrated and uncalibrated hyperspectral images from the AVIRIS and Hyperion sensors. The results provided by the proposed HyperLCA compressor have been evaluated and compared against those produced by the most relevant state-of-the-art compression solutions. The theoretical and experimental evidence indicates that the proposed algorithm represents an excellent option for lossy compressing hyperspectral images, especially for applications where the available computational resources are limited, such as on-board scenarios. Full article

With success of Deep Belief Networks (DBNs) in computer vision, DBN has attracted great attention in hyperspectral classification. Many deep learning based algorithms have been focused on deep feature extraction for classification improvement. Multi-features, such as texture feature, are widely utilized in classification process to enhance classification accuracy greatly. In this paper, a novel hyperspectral classification framework based on an optimal DBN and a novel texture feature enhancement (TFE) is proposed. Through band grouping, sample band selection and guided filtering, the texture features of hyperspectral data are improved. After TFE, the optimal DBN is employed on the hyperspectral reconstructed data for feature extraction and classification. Experimental results demonstrate that the proposed classification framework outperforms some state-of-the-art classification algorithms, and it can achieve outstanding hyperspectral classification performance. Furthermore, our proposed TFE method can play a significant role in improving classification accuracy. Full article

Fire impacts many vegetated ecosystems across the world. The severity of a fire is major component in determining post-fire effects, including soil erosion, trace gas emissions, and the trajectory of recovery. In this study, we used imaging spectroscopy data combined with Multiple Endmember Spectral Mixture Analysis (MESMA), a form of spectral mixture analysis that accounts for endmember variability, to map fire severity of the 2013 Rim Fire. We evaluated four endmember selection approaches: Iterative Endmember Selection (IES), count-based within endmember class (In-CoB), Endmember Average Root Mean Squared Error (EAR), and Minimum Average Spectral Angle (MASA). To reduce the dimensionality of the imaging spectroscopy data we used uncorrelated Stable Zone Unmixing (uSZU). Fractional cover maps derived from MESMA were validated using two approaches: (1) manual interpretation of fine spatial resolution WorldView-2 imagery; and (2) ground plots measuring the Geo Composite Burn Index (GeoCBI) and the percentage of co-dominant and dominant trees with green, brown, and black needles. Comparison to reference data demonstrated fairly high correlation for green vegetation and char fractions (r² values as high as 0.741 for the MESMA ash fractions compared to classified WorldView-2 imagery and as high as 0.841 for green vegetation fractions). The combination of uSZU band selection and In-CoB endmember selection had the best trade-off between accuracy and computational efficiency. This study demonstrated that detailed fire severity retrievals based on imaging spectroscopy can be optimized using techniques that would be viable also in a satellite-based imaging spectrometer. Full article

Restricted by technical and budget constraints, hyperspectral (HS) image which contains abundant spectral information generally has low spatial resolution. Fusion of hyperspectral and panchromatic (PAN) images can merge spectral information of the former and spatial information of the latter. In this paper, a new hyperspectral image fusion algorithm using structure tensor is proposed. An image enhancement approach is utilized to sharpen the spatial information of the PAN image, and the spatial details of the HS image is obtained by an adaptive weighted method. Since structure tensor represents structure and spatial information, a structure tensor is introduced to extract spatial details of the enhanced PAN image. Seeing that the HS and PAN images contain different and complementary spatial information for a same scene, a weighted fusion method is presented to integrate the extracted spatial information of the two images. To avoid artifacts at the boundaries, a guided filter is applied to the integrated spatial information image. The injection matrix is finally constructed to reduce spectral and spatial distortion, and the fused image is generated by injecting the complete spatial information. Comparative analyses validate the proposed method outperforms the state-of-art fusion methods, and provides more spatial details while preserving the spectral information. Full article

Band selection (BS) is one of the important topics in hyperspectral image (HSI) processing. Many types of BS algorithms were proposed in the last decade. However, most of them were designed for off-line use. They can only be used with pre-collected data, and are sometimes ineffective for applications that require timeliness, such as disaster prevention or target detection. This paper proposes an online BS method that allows us obtain instant BS results in a progressive manner during HSI data transmission, which is carried out under band-interleaved-by-sample/pixel (BIS/BIP) format. Such a revolutionary method is called progressive sample processing of band selection (PSP-BS). In PSP-BS, BS can be done recursively pixel by pixel, so that the instantaneous BS can be achieved without waiting for all the pixels of an image. To develop a PSP-BS algorithm, we proposed PSP-OMPBS, which adopted the recursive version of a self-sparse regression BS method (OMPBS) as a native algorithm. The experiments conducted on two real hyperspectral images demonstrate that PSP-OMPBS can progressively output the BS with extremely low computing time. In addition, the convergence of BS results during transmission can be further accelerated by using a pre-defined pixel transmission sequence. Such a significant advantage not only allows BS to be done in a real-time manner for the future satellite data downlink, but also determines the BS results in advance, without waiting to receive every pixel of an image. Full article

The detection of water stress in vineyards plays an integral role in the sustainability of high-quality grapes and prevention of devastating crop loses. Hyperspectral remote sensing technologies combined with machine learning provides a practical means for modelling vineyard water stress. In this study, we applied two ensemble learners, i.e., random forest (RF) and extreme gradient boosting (XGBoost), for discriminating stressed and non-stressed Shiraz vines using terrestrial hyperspectral imaging. Additionally, we evaluated the utility of a spectral subset of wavebands, derived using RF mean decrease accuracy (MDA) and XGBoost gain. Our results show that both ensemble learners can effectively analyse the hyperspectral data. When using all wavebands (p = 176), RF produced a test accuracy of 83.3% (KHAT (kappa analysis) = 0.67), and XGBoost a test accuracy of 80.0% (KHAT = 0.6). Using the subset of wavebands (p = 18) produced slight increases in accuracy ranging from 1.7% to 5.5% for both RF and XGBoost. We further investigated the effect of smoothing the spectral data using the Savitzky-Golay filter. The results indicated that the Savitzky-Golay filter reduced model accuracies (ranging from 0.7% to 3.3%). The results demonstrate the feasibility of terrestrial hyperspectral imagery and machine learning to create a semi-automated framework for vineyard water stress modelling. Full article

Two wide-swath hyperspectral imaging microsatellites, SPARK-01 and -02, were launched on 22 December 2016. Radiometric calibration coefficients were determined for these two satellites via a calibration experiment performed from the end of February to the beginning of March 2017 at the high-altitude, homogenous Dunhuang calibration site in the Gobi Desert in China. In-situ measurements, including ground reflectance, direct transmittance, diffuse-to-global irradiance ratio, and radiosonde vertical profile, were acquired. A unique relative calibration procedure was developed using actual satellite images. This procedure included dark current computation and non-uniform correction processes. The former was computed by averaging multiple lines of long strip imagery acquired over open oceans during nighttime, while the latter was computed using images acquired after the adjustment of the satellite yaw angle to 90°. This technique was shown to be suitable for large-swath satellite image relative calibration. After relative calibration, reflectance, irradiance, and improved irradiance-based methods were used to conduct absolute radiometric calibrations in order to predict the top-of-atmosphere (TOA) radiance. The SPARK-01 and -02 satellites passed over the calibration site on 7 March and 28 February 2017, during which time fair and non-ideal weather occurred, respectively. Thus, the SPARK-01 calibration coefficient was derived using reflectance- and irradiance-based methods, while that of SPARK -02 was derived using reflectance- and improved irradiance-based methods. The sources of calibration uncertainty, which include aerosol-type assumptions, transmittance measurements, water vapor content retrieval, spectral wavelength shift and satellite image misregistration, were explored in detail for different calibration methods. Using the reflectance and irradiance-based methods, the total uncertainty for SPARK-01 was estimated to be 4.7% and 4.1%, respectively, in the <1000 nm spectral range. For SPARK-02, total uncertainties of 8.1% and of 5.9% were estimated using the reflectance- and improved irradiance-based methods, respectively. The calibration methods were also verified using MODIS images, which confirmed that the calibration accuracies were within the expected range. These in-situ measurements, analyses, and results provide a basis for in-orbit radiometric calibration of the SPARK-01 and -02 satellites. These experiments strongly support the use of diffuse-to-global ratio measurements in in-situ vicarious calibration experiments and the addition of spectrally continuous measurements for direct transmittance, which is important for hyperspectral satellite sensors. Full article

This paper develops a new approach to band subset selection (BSS) for hyperspectral image classification (HSIC) which selects multiple bands simultaneously as a band subset, referred to as simultaneous multiple band selection (SMMBS), rather than one band at a time sequentially, referred to as sequential multiple band selection (SQMBS), as most traditional band selection methods do. In doing so, a criterion is particularly developed for BSS that can be used for HSIC. It is a linearly constrained minimum variance (LCMV) derived from adaptive beamforming in array signal processing which can be used to model misclassification errors as the minimum variance. To avoid an exhaustive search for all possible band subsets, two numerical algorithms, referred to as sequential (SQ) and successive (SC) algorithms are also developed for LCMV-based SMMBS, called SQ LCMV-BSS and SC LCMV-BSS. Experimental results demonstrate that LCMV-based BSS has advantages over SQMBS. Full article

Anomaly detection has received considerable interest for hyperspectral data exploitation due to its high spectral resolution. Fast processing and good detection performance are practically significant in real world problems. Aiming at these requirements, this paper develops a recursive local summation RX anomaly detection approach by virtue of sliding windows. This paper develops a recursive local summation RX anomaly detection approach by virtue of sliding windows. A causal sample covariance/correlation matrix is derived for local window background. As for the real-time sliding windows, the $Woodbury$ identity is used in recursive update equations, which could avoid the calculation of historical information and thus speed up the processing. Furthermore, a background suppression algorithm is also proposed in this paper, which removes the current under test pixel from the recursively update processing. Experiments are implemented on a real hyperspectral image. The experiment results demonstrate that the proposed anomaly detector outperforms the traditional real-time local background detector and has a significant speed-up effect on calculation time compared with the traditional detectors. Full article

Leaf maturation from initiation to senescence is a phenological event of plants that results from the influences of temperature and water availability on physiological activities during a life cycle. Detection of newly grown leaves (NGL) is therefore useful for the diagnosis of tree growth, tree stress, and even climatic change. This paper applies Constrained Energy Minimization (CEM), which is a hyperspectral target detection technique to spot grown leaves in a UAV multispectral image. According to the proportion of NGL in different regions, this paper proposes three innovative CEM based detectors: Subset CEM, Sliding Window-based CEM (SW CEM), and Adaptive Sliding Window-based CEM (AWS CEM). AWS CEM can especially adjust the window size according to the proportion of NGL around the current pixel. The results show that AWS CEM improves the accuracy of NGL detection and also reduces the false alarm rate. In addition, the results of the supervised target detection depend on the appropriate signature. In this case, we propose the Optimal Signature Generation Process (OSGP) to extract the optimal signature. The experimental results illustrate that OSGP can effectively improve the stability and the detection rate. Full article

Sparse unmixing is widely used for hyperspectral imagery to estimate the optimal fraction (abundance) of materials contained in mixed pixels (endmembers) of a hyperspectral scene, by considering the abundance sparsity. This abundance has a unique property, i.e., high spatial correlation in local regions. This is due to the fact that the endmembers existing in the region are highly correlated. This implies the low-rankness of the abundance in terms of the endmember. From this prior knowledge, it is expected that considering the low-rank local abundance to the sparse unmixing problem improves estimation performance. In this study, we propose an algorithm that exploits the low-rank local abundance by applying the nuclear norm to the abundance matrix for local regions of spatial and abundance domains. In our optimization problem, the local abundance regularizer is collaborated with the $L_{2,1}$ norm and the total variation for sparsity and spatial information, respectively. We conducted experiments for real and simulated hyperspectral data sets assuming with and without the presence of pure pixels. The experiments showed that our algorithm yields competitive results and performs better than the conventional algorithms. Full article

White matter hyperintensities (WMHs) are closely related to various geriatric disorders including cerebrovascular diseases, cardiovascular diseases, dementia, and psychiatric disorders of elderly people, and can be generally detected on T2 weighted (T2W) or fluid attenuation inversion recovery (FLAIR) brain magnetic resonance (MR) images. This paper develops a new approach to detect WMH in MR brain images from a hyperspectral imaging perspective. To take advantage of hyperspectral imaging, a nonlinear band expansion (NBE) process is proposed to expand MR images to a hyperspectral image. It then redesigns the well-known hyperspectral subpixel target detection, called constrained energy minimization (CEM), as an iterative version of CEM (ICEM) for WMH detection. Its idea is to implement CEM iteratively by feeding back Gaussian filtered CEM-detection maps to capture spatial information. To show effectiveness of NBE-ICEM in WMH detection, the lesion segmentation tool (LST), which is an open source toolbox for statistical parametric mapping (SPM), is used for comparative study. For quantitative analysis, the synthetic images in BrainWeb provided by McGill University are used for experiments where our proposed NBE-ICEM performs better than LST in all cases, especially for noisy MR images. As for real images collected by Taichung Veterans General Hospital, the NBE-ICEM also shows its advantages over and superiority to LST. Full article

Integrating spectral and spatial information is proved effective in improving the accuracy of hyperspectral imagery classification. In recent studies, two kinds of approaches are widely investigated: (1) developing a multiple feature fusion (MFF) strategy; and (2) designing a powerful spectral-spatial feature extraction (FE) algorithm. In this paper, we combine the advantages of MFF and FE, and propose an ensemble based feature representation method for hyperspectral imagery classification, which aims at generating a hierarchical feature representation for the original hyperspectral data. The proposed method is composed of three cascaded layers: firstly, multiple features, including local, global and spectral, are extracted from the hyperspectral data. Next, a new hashing based feature representation method is proposed and conducted on the features obtained in the first layer. Finally, a simple but efficient extreme learning machine classifier is employed to get the classification results. To some extent, the proposed method is a combination of MFF and FE: instead of feature fusion or single feature extraction, we use an ensemble strategy to provide a hierarchical feature representation for the hyperspectral data. In the experiments, we select two popular and one challenging hyperspectral data sets for evaluation, and six recently proposed methods are compared. The proposed method achieves respectively 89.55%, 99.36% and 77.90% overall accuracies in the three data sets with 20 training samples per class. The results prove that the performance of the proposed method is superior to some MFF and FE based ones. Full article

Hyperspectral unmixing aims to estimate a set of endmembers and corresponding abundances in pixels. Nonnegative matrix factorization (NMF) and its extensions with various constraints have been widely applied to hyperspectral unmixing. $L_{1/2}$ and $L_2$ regularizers can be added to NMF to enforce sparseness and evenness, respectively. In practice, a region in a hyperspectral image may possess different sparsity levels across locations. The problem remains as to how to impose constraints accordingly when the level of sparsity varies. We propose a novel nonnegative matrix factorization with data-guided constraints (DGC-NMF). The DGC-NMF imposes on the unknown abundance vector of each pixel with either an $L_{1/2}$ constraint or an $L_2$ constraint according to its estimated mixture level. Experiments on the synthetic data and real hyperspectral data validate the proposed algorithm. Full article

Merging hyperspectral data from optical and thermal ranges allows a wider variety of minerals to be mapped and thus allows lithology to be mapped in a more complex way. In contrast, in most of the studies that have taken advantage of the data from the visible (VIS), near-infrared (NIR), shortwave infrared (SWIR) and longwave infrared (LWIR) spectral ranges, these different spectral ranges were analysed and interpreted separately. This limits the complexity of the final interpretation. In this study a presentation is made of how multiple absorption features, which are directly linked to the mineral composition and are present throughout the VIS, NIR, SWIR and LWIR ranges, can be automatically derived and, moreover, how these new datasets can be successfully used for mineral/lithology mapping. The biggest advantage of this approach is that it overcomes the issue of prior definition of endmembers, which is a requested routine employed in all widely used spectral mapping techniques. In this study, two different airborne image datasets were analysed, HyMap (VIS/NIR/SWIR image data) and Airborne Hyperspectral Scanner (AHS, LWIR image data). Both datasets were acquired over the Sokolov lignite open-cast mines in the Czech Republic. It is further demonstrated that even in this case, when the absorption feature information derived from multispectral LWIR data is integrated with the absorption feature information derived from hyperspectral VIS/NIR/SWIR data, an important improvement in terms of more complex mineral mapping is achieved. Full article

Ensemble learning is widely used to combine varieties of weak learners in order to generate a relatively stronger learner by reducing either the bias or the variance of the individual learners. Rotation forest (RoF), combining feature extraction and classifier ensembles, has been successfully applied to hyperspectral (HS) image classification by promoting the diversity of base classifiers since last decade. Generally, RoF uses principal component analysis (PCA) as the rotation tool, which is commonly acknowledged as an unsupervised feature extraction method, and does not consider the discriminative information about classes. Sometimes, however, it turns out to be sub-optimal for classification tasks. Therefore, in this paper, we propose an improved RoF algorithm, in which semi-supervised local discriminant analysis is used as the feature rotation tool. The proposed algorithm, named semi-supervised rotation forest (SSRoF), aims to take advantage of both the discriminative information and local structural information provided by the limited labeled and massive unlabeled samples, thus providing better class separability for subsequent classifications. In order to promote the diversity of features, we also adjust the semi-supervised local discriminant analysis into a weighted form, which can balance the contributions of labeled and unlabeled samples. Experiments on several hyperspectral images demonstrate the effectiveness of our proposed algorithm compared with several state-of-the-art ensemble learning approaches. Full article

Variable environmental conditions cause different spectral responses of scene endmembers. Ignoring these variations affects the accuracy of fractional abundances obtained from linear spectral unmixing. On the other hand, the correlation between the bands of hyperspectral data is not considered by conventional methods developed for dealing with spectral variability. In this paper, a novel approach is proposed to simultaneously mitigate spectral variability and reduce correlation among different endmembers in hyperspectral datasets. The idea of the proposed method is to utilize the angular discrepancy of bands in the Prototype Space (PS), which is constructed using the endmembers of the image. Using the concepts of PS, in which each band is treated as a space point, we proposed a method to identify independent bands according to their angles. The proposed method comprised two main steps. In the first step, which aims to alleviate the spectral variability issue, image bands are prioritized based on their standard deviations computed over some sets of endmembers. Independent bands are then recognized in the prototype space, employing the angles between the prioritized bands. Finally, the unmixing process is done using the selected bands. In addition, the paper presents a technique to form a spectral library of endmembers’ variability (sets of endmembers). The proposed method extracts endmembers sets directly from the image data via a modified version of unsupervised spatial–spectral preprocessing. The performance of the proposed method was evaluated by five simulated images and three real hyperspectral datasets. The experiments show that the proposed method—using both groups of spectral variability reduction methods and independent band selection methods—produces better results compared to the conventional methods of each group. The improvement in the performance of the proposed method is observed in terms of more appropriate bands being selected and more accurate fractional abundance values being estimated. Full article

Airborne imaging spectroscopy (IS) and laser scanning (ALS) have been explored widely for tree species classification during the past decades. However, African agroforestry areas, where a few exotic tree species are dominant and many native species occur less frequently, have not yet been studied. Obtaining maps of tree species would provide useful information for the characterization of agroforestry systems and detecting invasive species. Our objective was to study tree species classification in a diverse tropical landscape using IS and ALS data at the tree crown level, with primary interest in the exotic tree species. We performed multiple analyses based on different IS and ALS feature sets, identified important features using feature selection, and evaluated the impact of combining the two data sources. Given that a high number of tree species with limited sample size (499 samples for 31 species) was expected to limit the classification accuracy, we tested different approaches to group the species based on the frequency of their occurrence and Jeffries–Matusita (JM) distance. Surface reflectance at wavelengths between 400–450 nm and 750–800 nm, and height to crown width ratio, were identified as important features. Nonetheless, a selection of minimum noise fraction (MNF) transformed reflectance bands showed superior performance. Support vector machine classifier performed slightly better than the random forest classifier, but the improvement was not statistically significant for the best performing feature set. The highest F1-scores were achieved when each of the species was classified separately against a mixed group of all other species, which makes this approach suitable for invasive species detection. Our results are valuable for organizations working on biodiversity conservation and improving agroforestry practices, as we showed how the non-native Eucalyptus spp., Acacia mearnsii and Grevillea robusta (mean F1-scores 76%, 79% and 89%, respectively) trees can be mapped with good accuracy. We also found a group of six fruit bearing trees using JM distance, which was classified with mean F1-score of 65%. This was a useful finding, as these species could not be classified with acceptable accuracy individually, while they all share common economic and ecological importance. Full article

Sparse Representation has been widely applied to classification of hyperspectral images (HSIs). Besides spectral information, the spatial context in HSIs also plays an important role in the classification. The recently published Multiscale Adaptive Sparse Representation (MASR) classifier has shown good performance in exploiting spatial information for HSI classification. But the spatial information is exploited by multiscale patches with fixed sizes of square windows. The patch can include all nearest neighbor pixels but these neighbor pixels may contain some noise pixels. Then another research proposed a Multiscale Superpixel-Based Sparse Representation (MSSR) classifier. Shape-adaptive superpixels can provide more accurate representation than patches. But it is difficult to select scales for superpixels. Therefore, inspired by the merits and demerits of multiscale patches and superpixels, we propose a novel algorithm called Multiscale Union Regions Adaptive Sparse Representation (MURASR). The union region, which is the overlap of patch and superpixel, can make full use of the advantages of both and overcome the weaknesses of each one. Experiments on several HSI datasets demonstrate that the proposed MURASR is superior to MASR and union region is better than the patch in the sparse representation. Full article

This study aims to evaluate three classes of methods to discriminate between 13 peatland vegetation types using reflectance data. These vegetation types were empirically defined according to their composition, strata and biodiversity richness. On one hand, it is assumed that the same vegetation type spectral signatures have similarities. Consequently, they can be compared to a reference spectral database. To catch those similarities, several similarities criteria (related to distances (Euclidean distance, Manhattan distance, Canberra distance) or spectral shapes (Spectral Angle Mapper) or probabilistic behaviour (Spectral Information Divergence)) and several mathematical transformations of spectral signatures enhancing absorption features (such as the first derivative or the second derivative, the normalized spectral signature, the continuum removal, the continuum removal derivative reflectance, the log transformation) were investigated. Furthermore, those similarity measures were applied on spectral ranges which characterize specific biophysical properties. On the other hand, we suppose that specific biophysical properties/components may help to discriminate between vegetation types applying supervised classification such as Random Forest (RF), Support Vector Machines (SVM), Regularized Logistic Regression (RLR), Partial Least Squares-Discriminant Analysis (PLS-DA). Biophysical components can be used in a local way considering vegetation spectral indices or in a global way considering spectral ranges and transformed spectral signatures, as explained above. RLR classifier applied on spectral vegetation indices (training size = 25%) was able to achieve 77.21% overall accuracy in discriminating peatland vegetation types. It was also able to discriminate between 83.95% vegetation types considering specific spectral range [[range-phrase = –]3501350 $\mathrm{n}\mathrm{m}$ ], first derivative of spectral signatures and training size = 25%. Conversely, similarity criterion was able to achieve 81.70% overall accuracy using the Canberra distance computed on the full spectral range [[range-phrase = –]3502500 $\mathrm{n}\mathrm{m}$ ]. The results of this study suggest that RLR classifier and similarity criteria are promising to map the different vegetation types with high ecological values despite vegetation heterogeneity and mixture. Full article