Search

Many applications of hyperspectral remote sensing involve automatic detection of solid targets. The design and applicability of such systems depend on their detection performance under various deployment scenarios; thus, it is useful to develop an accurate and realistic performance model. In this paper, we present a hyperspectral target detection performance prediction model for the common matched filter and normalized matched filter detection algorithms. A statistical model for hyperspectral data is discussed, which includes a replacement model for target pixels and a finite mixture of $t$ -elliptically contoured distributions for background materials. For this general input model, analytic forms for detector output distributions are not always available. The main contribution of this paper is to develop an efficient and robust performance simulation technique using Monte Carlo (MC) with importance sampling. Compared to standard MC, our simulation technique requires a substantially smaller sample size to obtain accurate performance estimates at low false alarm rates. The proposed technique is useful for predicting detection performance over a wide range of input models when analytic solutions are unavailable. [ABSTRACT FROM AUTHOR]

In this paper, a new joint spectral–spatial subpixel mapping model is proposed for hyperspectral remotely sensed imagery. Conventional approaches generally use an intermediate step based on the derivation of fractional abundance maps obtained after a spectral unmixing process, and thus the rich spectral information contained in the original hyperspectral data set may not be utilized fully. In this paper, a concept of subpixel abundance map, which calculates the abundance fraction of each subpixel to belong to a given class, was introduced. This allows us to directly connect the original (coarser) hyperspectral image with the final subpixel result. Furthermore, the proposed approach incorporates the spectral information contained in the original hyperspectral imagery and the concept of spatial dependence to generate a final subpixel mapping result. The proposed approach has been experimentally evaluated using both synthetic and real hyperspectral images, and the obtained results demonstrate that the method achieves better results when compared to other seven subpixel mapping methods. The numerical comparisons are based on different indexes such as the overall accuracy and the CPU time. Moreover, the obtained results are statistically significant at 95% confidence. [ABSTRACT FROM AUTHOR]

This paper reports the development of a broadband highly reflective and structurally flat large-area 2-D subwavelength grating (SWG) reflector, based on a tensioned single-layer silicon metamaterial membrane. Rigorous coupled-wave analysis is adopted to design the metamaterial reflector, resulting in a wavelength range of 600 nm (1.9- $2.5~\mu \text{m}$) with >99% reflectivity, which represents the largest reported fractional bandwidth $\Delta \lambda /\lambda _{\mathrm {c}}$ of 27%. Effects of design parameter deviations on the SWG reflector reflectivity are studied, showing that the reflector design has good fabrication tolerances. A freestanding 2-mm dimension 2-D SWG reflector has been fabricated with high precision, and optical measurements indicate polarization-independent average reflectivity of 99% over the wavelength range of 1.91- $2.49~\mu \text{m}$ , which agrees well with the modeled result. This paper also, for the first time, examines the surface flatness characteristics of a suspended single-layer 2-D SWG reflector. With a tensile stress of 10 ± 5 MPa in the silicon grating membrane, the freestanding reflector achieves nanometer-scale surface flatness, which provides a significant advantage over multilayer DBRs requiring stress-balancing. Furthermore, a hyperspectral Fabry-Perot shortwave infrared filter based on a top suspended 2-D SWG reflector and a bottom 4-pair Si/SiO2 distributed Bragg reflector (DBR) has been experimentally demonstrated for the first time. The filter is measured to have a peak transmission of 80% with a narrow full-width at half-maximum (FWHM) of about 4 nm. This spectral resolution is one order of magnitude higher than that of other reported MEMS/DBR-based Fabry-Perot filters and is well-suited to highly demanding hyperspectral shortwave infrared imaging applications. [2022-0060] [ABSTRACT FROM AUTHOR]

The papers in this special section focus on the technology and applications supported by hyperspectral imaging and signal processing. [ABSTRACT FROM PUBLISHER]

The localization of changes that occur between the images in a multitemporal series is crucial for many applications, ranging from environmental monitoring to military surveillance. In contrast to traditional change detection methods, unmixing-based change detection has been shown to have the important added benefit of providing subpixel-level information on the nature of the changes, instead of only providing the location of the changes. Recently, sparse unmixing has also been introduced to hyperspectral change detection, resulting in a method that circumvents the drawbacks of regular spectral unmixing approaches. Sparse unmixing-based change detection reveals the changes that occur in a multitemporal series, at subpixel level, and in terms of the library spectra and their sparse abundances, and provides enhanced change detection performance, especially when subpixel-level changes have occurred. However, sparse unmixing is generally an ill-conditioned and time-consuming process, especially as the size of the utilized spectral library increases. In this paper, dictionary pruning is exploited for the first time for hyperspectral change detection using sparse unmixing, in order to alleviate the ill-conditioning of the problem and achieve decreased computation times and enhanced change detection performance. Experimental results on both realistic synthetic and real datasets are used to validate the proposed approach. [ABSTRACT FROM PUBLISHER]

One of the critical advantages of a high-altitude airship (HAA) is its ability to carry multiple-task payloads for remote sensing monitoring on the plateau. We developed a multi-payload observation system based on the HAA (ASQ-MPHAAS) for remote sensing monitoring of the Qinghai plateau. In this paper, we demonstrate the design, development, implementation of the ASQ-MPHAAS. The ASQ-MPHAAS includes two parts; one part is our HAA (ASQ-HAA380), and the other part is our multi-payload observation system (ASQ-MPOS-1). The ASQ-MPOS-1 consists of two hyperspectral push-broom imaging sensors, a multispectral imaging sensor, a high-resolution RGB camera, a video sensor, a Position and Orientation System (POS), and other components. The ASQ-MPOS-1 is installed under the ASQ-HAA380. We also introduce some key methods of data processing we propose to solve unique problems brought by HAA. Four real-world field experiments are presented in this paper to demonstrate that the ASQ-MPHAAS can capture different kinds of remote sensing data at a high spatial resolution of a few or dozen centimeters. [ABSTRACT FROM AUTHOR]

Owing to the limits of incident energy and hardware system, hyperspectral (HS) images always suffer from low spatial resolution, compared with multispectral (MS) or panchromatic (PAN) images. Therefore, image fusion has emerged as a useful technology that is able to combine the characteristics of high spectral and spatial resolutions of HS and PAN/MS images. In this paper, a novel HS and PAN image fusion method based on convolutional neural network (CNN) is proposed. The proposed method incorporates the ideas of both hyper-sharpening and MS pan-sharpening techniques, thereby employing a two-stage cascaded CNN to reconstruct the anticipated high-resolution HS image. Technically, the proposed CNN architecture consists of two sub-networks, the detail injection sub-network and unmixing sub-network. The former aims at producing a latent high-resolution MS image, whereas the latter estimates the desired high-resolution abundance maps by exploring the spatial and spectral information of both HS and MS images. Moreover, two model-training fashions are presented in this paper for the sake of effectively training our network. Experiments on simulated and real remote sensing data demonstrate that the proposed method can improve the spatial resolution and spectral fidelity of HS image, and achieve better performance than some state-of-the-art HS pan-sharpening algorithms. [ABSTRACT FROM AUTHOR]

Real-time hyperspectral imaging on-board compression represents a critical processing step in many remote sensing applications where the acquired hyperspectral data need to be efficiently stored and/or transferred. However, the complexity of the compression algorithms as well as the volume of data to be compressed and the limited computational resources of the hardware devices available on-board turn the real-time compression into a very challenging task. This paper presents a low-power-consumption solution for real-time lossy compression of hyperspectral images. The lossy compression algorithm for hyperspectral image system (HyperLCA) compressor has been implemented onto two NVIDIA Jetson developer kits. These NVIDIA boards include low-power embedded graphic processing units, which allow parallel programing for speeding up the compression process at a reasonable low power consumption. The experiments carried out in this paper are oriented to the necessities imposed by a specific smart farming application although all drawn conclusions are extrapolable to other fields in which remotely sensed hyperspectral images are to be compressed in real time. The obtained results verify both the good performance of the HyperLCA compressor for the targeted application and the achievement of a real-time performance by using the developed implementations. Additionally, several comparisons and conclusions have been drawn from the experiments in relation to the different strategies employed for accelerating the compression process. [ABSTRACT FROM AUTHOR]

Recently, researchers have shown the powerful ability of deep methods with multilayers to extract high-level features and to obtain better performance for hyperspectral image classification. However, a common problem of traditional deep models is that the learned deep models might be suboptimal because of the limited number of training samples, especially for the image with large intraclass variance and low interclass variance. In this paper, novel convolutional neural networks (CNNs) with multiscale convolution (MS-CNNs) are proposed to address this problem by extracting deep multiscale features from the hyperspectral image. Moreover, deep metrics usually accompany with MS-CNNs to improve the representational ability for the hyperspectral image. However, the usual metric learning would make the metric parameters in the learned model tend to behave similarly. This similarity leads to obvious model’s redundancy and, thus, shows negative effects on the description ability of the deep metrics. Traditionally, determinantal point process (DPP) priors, which encourage the learned factors to repulse from one another, can be imposed over these factors to diversify them. Taking advantage of both the MS-CNNs and DPP-based diversity-promoting deep metrics, this paper develops a CNN with multiscale convolution and diversified metric to obtain discriminative features for hyperspectral image classification. Experiments are conducted over four real-world hyperspectral image data sets to show the effectiveness and applicability of the proposed method. Experimental results show that our method is better than original deep models and can produce comparable or even better classification performance in different hyperspectral image data sets with respect to spectral and spectral–spatial features. [ABSTRACT FROM AUTHOR]

Hyperspectral imagery (HSI) has shown promising results in real-world applications. However, the technological evolution of optical sensors poses two main challenges in HSI classification: 1) the spectral band is usually redundant and noisy and 2) HSI with millions of pixels has become increasingly common in real-world applications. Motivated by the recent success of hybrid huberized support vector machines (HHSVMs), which inherit the benefits of both lasso and ridge regression, this paper first investigates the advantages of HHSVM for HSI applications. Unfortunately, the existing HHSVM solvers suffer from prohibitive computational costs on large-scale data sets. To solve this problem, this paper proposes simple and effective stochastic HHSVM algorithms for HSI classification. In the stochastic settings, we show that with a probability of at least $1-\varrho $ , our algorithms find an $\epsilon $ -accurate solution using $\tilde {O}({1}/{\lambda _{2}\epsilon })$ iterations. Since the convergence rate of our algorithms does not depend on the size of the training set, our algorithms are suitable for handling large-scale problems. We demonstrate the superiority of our algorithms by conducting experiments on large-scale binary and multiclass classification problems, comparing to the state-of-the-art HHSVM solvers. Finally, we apply our algorithms to real HSI classification and achieve promising results. [ABSTRACT FROM AUTHOR]

Hyperspectral imaging is a widely used technique in remote sensing in which an imaging spectrometer collects hundreds of images (at different wavelength channels) for the same area on the surface of the earth. In the last two decades, several methods (unsupervised, supervised, and semisupervised) have been proposed to deal with the hyperspectral image classification problem. Supervised techniques have been generally more popular, despite the fact that it is difficult to collect labeled samples in real scenarios. In particular, deep neural networks, such as convolutional neural networks (CNNs), have recently shown a great potential to yield high performance in the hyperspectral image classification. However, these techniques require sufficient labeled samples in order to perform properly and generalize well. Obtaining labeled data is expensive and time consuming, and the high dimensionality of hyperspectral data makes it difficult to design classifiers based on limited samples (for instance, CNNs overfit quickly with small training sets). Active learning (AL) can deal with this problem by training the model with a small set of labeled samples that is reinforced by the acquisition of new unlabeled samples. In this paper, we develop a new AL-guided classification model that exploits both the spectral information and the spatial-contextual information in the hyperspectral data. The proposed model makes use of recently developed Bayesian CNNs. Our newly developed technique provides robust classification results when compared with other state-of-the-art techniques for hyperspectral image classification. [ABSTRACT FROM AUTHOR]

This paper presents a novel approach for spectral unmixing of remotely sensed hyperspectral data. It exploits probabilistic latent topics in order to take advantage of the semantics pervading the latent topic space when identifying spectral signatures and estimating fractional abundances from hyperspectral images. Despite the contrasted potential of topic models to uncover image semantics, they have been merely used in hyperspectral unmixing as a straightforward data decomposition process. This limits their actual capabilities to provide semantic representations of the spectral data. The proposed model, called dual-depth sparse probabilistic latent semantic analysis (DEpLSA), makes use of two different levels of topics to exploit the semantic patterns extracted from the initial spectral space in order to relieve the ill-posed nature of the unmixing problem. In other words, DEpLSA defines a first level of deep topics to capture the semantic representations of the spectra, and a second level of restricted topics to estimate endmembers and abundances over this semantic space. An experimental comparison in conducted using the two standard topic models and the seven state-of-the-art unmixing methods available in the literature. Our experiments, conducted using four different hyperspectral images, reveal that the proposed approach is able to provide competitive advantages over available unmixing approaches. [ABSTRACT FROM AUTHOR]

To solve the low spatial and/or temporal resolution problem which the conventional hyperspectral cameras often suffer from, coded hyperspectral imaging systems have attracted more attention recently. Recovering a hyperspectral image (HSI) from its corresponding coded image is an ill-posed inverse problem, and learning accurate prior of HSI is essential to solve this inverse problem. In this paper, we present an effective convolutional neural network (CNN) based method for coded HSI reconstruction, which learns the deep prior from the external dataset as well as the internal information of input coded image with spatial-spectral constraint. Specifically, we first develop a CNN-based channel attention reconstruction network to effectively exploit the spatial-spectral correlation of the HSI. Then, the reconstruction network is learned by leveraging an arbitrary external hyperspectral dataset to exploit the general spatial-spectral correlation under adversarial loss. Finally, we customize the network by internal learning with spatial-spectral constraint and total variation regularization for each coded image, which can make use of the internal imaging model to learn specific prior for current desirable image and effectively avoids overfitting. Experimental results using both synthetic data and real images show that our method outperforms the state-of-the-art methods on several popular coded hyperspectral imaging systems under both comprehensive quantitative metrics and perceptive quality. [ABSTRACT FROM AUTHOR]

Hyperspectral image produces high spectral resolution at the sacrifice of spatial resolution. Without reducing the spectral resolution, improving the resolution in the spatial domain is a very challenging problem. Motivated by the discovery that hyperspectral image exhibits high similarity between adjacent bands in a large spectral range, in this paper, we explore a new structure for hyperspectral image super-resolution (DualSR), leading to a dual-stage design, i.e., coarse stage and fine stage. In coarse stage, five bands with high similarity in a certain spectral range are divided into three groups, and the current band is guided to study the potential knowledge. Under the action of alternative spectral fusion mechanism, the coarse SR image is super-resolved in band-by-band. In order to build model from a global perspective, an enhanced back-projection method via spectral angle constraint is developed in fine stage to learn the content of spatial-spectral consistency, dramatically improving the performance gain. Extensive experiments demonstrate the effectiveness of the proposed coarse stage and fine stage. Besides, our network produces state-of-the-art results against existing works in terms of spatial reconstruction and spectral fidelity. Our code is publicly available at https://github.com/qianngli/DualSR. [ABSTRACT FROM AUTHOR]

High spatial resolution and high spectral resolution images (HR-HSIs) are widely applied in geosciences, medical diagnosis, and beyond. However, how to get images with both high spatial resolution and high spectral resolution is still a problem to be solved. In this paper, we present a deep spatial-spectral feature interaction network (SSFIN) for reconstructing an HR-HSI from a low-resolution multispectral image (LR-MSI), e.g., RGB image. In particular, we introduce two auxiliary tasks, i.e., spatial super-resolution (SR) and spectral SR to help the network recover the HR-HSI better. Since higher spatial resolution can provide more detailed information about image texture and structure, and richer spectrum can provide more attribute information, we propose a spatial-spectral feature interaction block (SSFIB) to make the spatial SR task and the spectral SR task benefit each other. Therefore, we can make full use of the rich spatial and spectral information extracted from the spatial SR task and spectral SR task, respectively. Moreover, we use a weight decay strategy (for the spatial and spectral SR tasks) to train the SSFIN, so that the model can gradually shift attention from the auxiliary tasks to the primary task. Both quantitative and visual results on three widely used HSI datasets demonstrate that the proposed method achieves a considerable gain compared to other state-of-the-art methods. Source code is available at https://github.com/junjun-jiang/SSFIN. [ABSTRACT FROM AUTHOR]

Convolutional Neural Networks (CNN) and Graph Neural Networks (GNN), such as Graph Attention Networks (GAT), are two classic neural network models, which are applied to the processing of grid data and graph data respectively. They have achieved outstanding performance in hyperspectral images (HSIs) classification field, which have attracted great interest. However, CNN has been facing the problem of small samples and GNN has to pay a huge computational cost, which restrict the performance of the two models. In this paper, we propose Weighted Feature Fusion of Convolutional Neural Network and Graph Attention Network (WFCG) for HSI classification, by using the characteristics of superpixel-based GAT and pixel-based CNN, which proved to be complementary. We first establish GAT with the help of superpixel-based encoder and decoder modules. Then we combined the attention mechanism to construct CNN. Finally, the features are weighted fusion with the characteristics of two neural network models. Rigorous experiments on three real-world HSI data sets show WFCG can fully explore the high-dimensional feature of HSI, and obtain competitive results compared to other state-of-the art methods. [ABSTRACT FROM AUTHOR]

Hyperspectral imagery with very high spectral resolution provides a new insight for subtle nuances identification of similar substances. However, hyperspectral target detection faces significant challenges of intraclass dissimilarity and interclass similarity due to the unavoidable interference caused by atmosphere, illumination, and sensor noise. In order to effectively alleviate these spectral inconsistencies, this paper proposes a novel target detection method without strict assumptions on data distribution based on an unconstrained linear mixture model and deep learning. Our proposed detector firstly reduces interference via a specifically designed deep-learning-based hierarchical denoising autoencoder, and then carries out accurate detection with a two-step subspace projection, aiming at background suppression and target enhancement. Additionally, to generate representative background and reliable target samples required in the detection procedure, an efficient spatial-spectral unified endmember extraction method has been developed. Performance comparison with several state-of-the-art detection methods and further analysis on four real-world hyperspectral images demonstrate the effectiveness and efficiency of our proposed target detector. [ABSTRACT FROM AUTHOR]

This paper develops a novel hyperspectral image (HSI) classification framework by exploiting the spectral-spatial features of multiscale superpixels via recurrent neural networks with stacked autoencoders. The superpixels can be used to segment an HSI into shape-adaptive regions, and multiscale superpixels can capture the object information more accurately. Therefore, the superpixel-based classification methods have been studied by many researchers. In this paper, we propose a multiscale superpixel-based classification method. In contrast to current research, the proposed method not only captures the features of each scale but also considers the correlation among different scales via recurrent neural networks. In this way, the spectral-spatial information within a superpixel is more efficiently exploited. In this paper, we first segment the HSI from coarse to fine scales using the superpixels. Then, the spatial features within each superpixel and among superpixels are sufficiently exploited by the local and nonlocal similarity measure. Finally, recurrent neural networks with stacked autoencoders are proposed to learn the high-level multiscale spectral-spatial features. Experiments are conducted on real HSI datasets. The results demonstrate the superiority of the proposed method over several well-known methods in both visual appearance and classification accuracy. [ABSTRACT FROM AUTHOR]

Compression of hyperspectral images onboard of spacecrafts is a tradeoff between the limited computational resources and the ever-growing spatial and spectral resolution of the optical instruments. As such, it requires low-complexity algorithms with good rate-distortion performance and high throughput. In recent years, the Consultative Committee for Space Data Systems (CCSDS) has focused on lossless and near-lossless compression approaches based on predictive coding, resulting in the recently published CCSDS 123.0-B-2 recommended standard. While the in-loop reconstruction of quantized prediction residuals provides excellent rate-distortion performance for the near-lossless operating mode, it significantly constrains the achievable throughput due to data dependencies. In this paper, we study the performance of a faster method based on the prequantization of the image followed by a lossless predictive compressor. While this is well known to be suboptimal, one can exploit powerful signal models to reconstruct the image at the ground segment, recovering part of the suboptimality. In particular, we show that convolutional neural networks can be used for this task and that they can recover the whole SNR drop incurred at a bit rate of 2 bits per pixel. [ABSTRACT FROM AUTHOR]

Double-clad fibers (DCFs) are versatile waveguides comprising a single-mode core surrounded by a multimode inner cladding. DCFs are increasingly used for multimodal biomedical applications, such as imaging or therapy, for which the core is typically used for coherent illumination and the inner cladding to support a concurrent modality. Double-clad fiber couplers (DCFCs) were developed to allow the separation or combination of these signals in a robust and efficient fashion. In this paper, we review the operation and types of DCFCs in addition to current biomedical applications. [ABSTRACT FROM AUTHOR]

In this paper, we propose a spectral–spatial unified network (SSUN) with an end-to-end architecture for the hyperspectral image (HSI) classification. Different from traditional spectral–spatial classification frameworks where the spectral feature extraction (FE), spatial FE, and classifier training are separated, these processes are integrated into a unified network in our model. In this way, both FE and classifier training will share a uniform objective function and all the parameters in the network can be optimized at the same time. In the implementation of the SSUN, we propose a band grouping-based long short-term memory model and a multiscale convolutional neural network as the spectral and spatial feature extractors, respectively. In the experiments, three benchmark HSIs are utilized to evaluate the performance of the proposed method. The experimental results demonstrate that the SSUN can yield a competitive performance compared with existing methods. [ABSTRACT FROM AUTHOR]

Recently, multifeature learning in collaborative representation classification (CRC) for hyperspectral images has generated promising performance. In this paper, two novel multifeature learning algorithms that update dictionary directly and indirectly are proposed. In order to offer the complementarity of multifeature, four different types of features—global feature (i.e., Gabor feature), local feature (i.e., local binary pattern), shape feature (i.e., extended multiattribute profiles), and spectral feature—are adopted in this paper. Under the hypothesis that most of the features should share the same coding pattern in CRC, this paper proposes to learn proper dictionaries for each feature until obtaining stable codes in a linear classifier. Furthermore, to avoid the explicit mapping of infinite-dimensional dictionaries in a nonlinear kernelized classifier, an indirect approach to construct the transformation matrix from original dictionaries to learn new dictionaries is developed. Three real hyperspectral images acquired from different sensors are adopted for performance evaluation. The experimental results demonstrate that the proposed methods can provide superior performance compared with those of the state-of-the-art classifiers. [ABSTRACT FROM AUTHOR]

The twenty six papers in this special issue focus on the technologies of hyperspectral remote sensing (HRS)and imaging spectroscopy. HRS has emerged as a powerful tool to understand phenomena at local and global scales by virtue of imaging through a diverse range of platforms, including terrestrial in-situ imaging platforms, unmanned and manned aerial vehicles, and satellite platforms. By virtue of imaging over a wide range of spectral wavelengths, it is possible to characterize object specific properties very accurately. As a result, hyperspectral imaging (also known as imaging spectroscopy) has gained popularity for a wide variety of applications, including environment monitoring, precision agriculture, mineralogy, forestry, urban planning, and defense applications. The increased analysis capability comes at a cost?there are a variety of challenges that must be overcome for robust image analysis of such data, including high dimensionality, limited sample size for training supervised models, noise and atmospheric affects, mixed pixels, etc. The papers in this issue represent some of the recent developments in image analysis algorithms and unique applications of hyperspectral imaging data. [ABSTRACT FROM PUBLISHER]

In order to provide reliable information about the instantaneous field of view considered in hyperspectral images through spectral unmixing, understanding the kind of mixture that occurs over each pixel plays a crucial role. In this paper, in order to detect nonlinear mixtures, a method for fast identification of linear mixtures is introduced. The proposed method does not need statistical information and performs an a priori test on the spectral linearity of each pixel. It uses standard least squares optimization to achieve estimates of the likelihood of occurrence of linear combinations of endmembers by taking advantage of the geometrical properties of hyperspectral signatures. Experimental results on both real and synthetic data sets show that the aforesaid algorithm is actually able to deliver a reliable and thorough assessment of the kind of mixtures present in the pixels of the scene. [ABSTRACT FROM PUBLISHER]

The classification accuracy of remote sensing data can be increased by integrating ancillary data provided by multisource acquisition of the same scene. We propose to merge the spectral and spatial content of hyperspectral images (HSIs) with elevation information from light detection and ranging (LiDAR) measurements. In this paper, we propose to fuse the data sets using orthogonal total variation component analysis (OTVCA). Extinction profiles are used to automatically extract spatial and elevation information from HSI and rasterized LiDAR features. The extracted spatial and elevation information is then fused with spectral information using the OTVCA-based feature fusion method to produce the final classification map. The extracted features have high dimension, and therefore OTVCA estimates the fused features in a lower dimensional space. OTVCA also promotes piece-wise smoothness while maintaining the spatial structures. Both attributes are important to provide homogeneous regions in the final classification maps. We benchmark the proposed approach (OTVCA-fusion) with an urban data set captured over an urban area in Houston/USA and a rural region acquired in Trento/Italy. In the experiments, OTVCA-fusion is evaluated using random forest and support vector machine classifiers. Our experiments demonstrate the ability of OTVCA-fusion to produce accurate classification maps while using fewer features compared with other approaches investigated in this paper. [ABSTRACT FROM PUBLISHER]

The AisaOWL is a recent-to-market thermal hyperspectral instrument. As such, there is little information about the sensor performance in the literature. The sensor covers the 7.6– 12.6~\mu \textm part of the long-wave infrared region with 102 continuous bands, and is capable of imaging in low-light conditions. This paper presents an independent characterization of the AisaOWL sensor, examining the spectral accuracy of black body measurements at different temperatures and validating manufacturer recommendations for warm-up, integration, and calibration times. This analysis is essential for establishing high quality operational procedures and in giving confidence to users of the data. In this paper, the sensor has been found to have a maximum error of 2 °C in absolute temperature measurement, and provides spectra most accurate in the 8– 9~\mu \textm region. The recommended warm-up time of 15 min has been confirmed, with a 1% increase in error identified for data collected only 7 min after switch on. The optimal integration time of 1.18 ms has been validated and an exponential decrease in performance observed outside the 0.85–1.2 ms range. The detector used by the sensor is shown to have stability issues and this has been examined by comparing black body data processed with different calibration data. While the detector is operating in a stable regime compatible with the calibration, these black body readings stay within 5% across the central bands, approaching 10% below 8~\mu \textm and just exceeding 20% above 11 \mu \textm . [ABSTRACT FROM AUTHOR]

Hyperspectral imaging (HSI) classification has become a popular research topic in recent years, and effective feature extraction is an important step before the classification task. Traditionally, spectral feature extraction techniques are applied to the HSI data cube directly. This paper presents a novel algorithm for HSI feature extraction by exploiting the curvelet-transformed domain via a relatively new spectral feature processing technique—singular spectrum analysis (SSA). Although the wavelet transform has been widely applied for HSI data analysis, the curvelet transform is employed in this paper since it is able to separate image geometric details and background noise effectively. Using the support vector machine classifier, experimental results have shown that features extracted by SSA on curvelet coefficients have better performance in terms of classification accuracy over features extracted on wavelet coefficients. Since the proposed approach mainly relies on SSA for feature extraction on the spectral dimension, it actually belongs to the spectral feature extraction category. Therefore, the proposed method has also been compared with some state-of-the-art spectral feature extraction techniques to show its efficacy. In addition, it has been proven that the proposed method is able to remove the undesirable artifacts introduced during the data acquisition process. By adding an extra spatial postprocessing step to the classified map achieved using the proposed approach, we have shown that the classification performance is comparable with several recent spectral–spatial classification methods. [ABSTRACT FROM PUBLISHER]

Nonlinear unmixing algorithms are playing a key role in modern earth observation analysis thanks to their ability to characterize complex phenomena occurring in the instantaneous field of view. When unmixing hyperspectral images according to nonlinear mixture models by means of state-of-the-art methods, actual abundances of the elements in the scene can be only indirectly estimated. Thus, the reliability of the investigation can be dramatically jeopardized, hence degrading the accuracy of the characterization of the surface composition. In order to overcome this issue, we propose in this paper a nonlinear programming scheme that aims at providing direct estimation of the end members fractions. The method we introduce is based on a structural optimization approach where the abundances are directly assessed, so that no epistemic uncertainties are injected in the framework. Experimental results show that the proposed method is able to deliver accurate and reliable estimates of these quantities in hyperspectral images. [ABSTRACT FROM AUTHOR]

Low spatial resolution hyperspectral (LRHS) and high spatial resolution multispectral (HRMS) image fusion has been recognized as an important technology for enhancing the spatial resolution of LRHS image. Recent advances in convolutional neural network have improved the performance of state-of-the-art fusion methods. However, it is still a challenging problem to effectively explore the spatial information of HRMS image. In this paper, we propose a pyramid fully convolutional network made up of an encoder sub-network and a pyramid fusion sub-network to address this issue. Specifically, the encoder sub-network aims to encode the LRHS image into a latent image. Then, this latent image, together with a HRMS image pyramid input, is used to progressively reconstruct the high spatial resolution hyperspectral image in a global-to-local manner. Furthermore, to sharpen the blurry predictions easily obtained by the standard $l_{2}$ loss, we introduce the gradient difference loss as a regularization term. We evaluate the proposed method on three datasets acquired by three different satellite sensors. Experimental results demonstrate that the proposed method achieves better performance than several state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Hyperspectral imagery (HSI) with a high spectral resolution contains hundreds and even thousands of spectral bands, and conveys abundant spectral information, which provides a unique advantage for target detection. A number of classical target detectors have been proposed based on the linear mixing model (LMM) and sparsity-based model. Compared with the LMM, sparsity-based detectors present a better performance on dealing with the spectral variability. Despite the great success of the sparsity-based model in recent years, one problem with all state-of-the-art sparsity-based models still exist: the target dictionary is formed via the target training samples that are selected from the global image scene. This is an improper way to construct target dictionary for hyperspectral target detection since the priori information is usually a given target spectrum obtained from a spectral library. Besides, target training samples selected from the global image scene are usually insufficient, which results in the problem that the target training samples and background training samples are unbalanced in the data volume, causing a deteriorated detection model. To tackle these problems, this paper constructs a target dictionary construction-based method, then proposes the constructed target dictionary-based sparsity-based target detection model and the constructed target dictionary-based sparse representation-based binary hypothesis model, which are called TDC-STD and TDC-SRBBH, respectively. Both of the proposed algorithms only need a given target spectrum as the input priori information. By using the given target spectrum for pre-detection via constrained energy minimization, we choose the pixels that have large output values as target training samples to construct the target dictionary. The proposed algorithms were tested on three benchmark HSI datasets and the experimental results show that the proposed algorithms demonstrate outstanding detection performances when compared with other state-of-the-art detectors. [ABSTRACT FROM AUTHOR]

Band selection is an effective approach to mitigate the “Hughes phenomenon” of hyperspectral image (HSI) classification. Recently, sparse representation (SR) theory has been successfully introduced to HSI band selection, and many SR-based methods have been developed and shown great potential and superiority. However, due to the inherent limitations of the SR scheme, i.e., individually representing each band with only a few other bands from the same subspace, the SR-based methods cannot effectively capture the global structures of the data, which limit the band selection performance. In this paper, to overcome this obstacle, the novel Laplacian-regularized low-rank subspace clustering (LLRSC) algorithm is proposed for HSI band selection. On the one hand, the low-rank subspace clustering model is introduced to capture the global structure information for the learned representation coefficient matrix and deal with the HSI band selection task in the clustering framework. On the other hand, considering the high correlation between adjacent bands, 1-D Laplacian regularization is utilized to incorporate the neighboring band information and further reduce the representation bias. Lastly, an eigenvalue analysis algorithm based on band mutation information is utilized to estimate the appropriate size of the band subset. The experimental results indicate that the proposed LLRSC algorithm outperforms the other state-of-the-art methods and achieves a very competitive band selection performance for HSIs. [ABSTRACT FROM AUTHOR]

Hyperspectral imagery (HSI) contains rich spectral information, which is beneficial to many tasks. However, acquiring HSI is difficult because of the limitations of current imaging technology. As an alternative method, spectral super-resolution aims at reconstructing HSI from its corresponding RGB image. Recently, deep learning has shown its power to this task, but most of the used networks are transferred from other domains, such as spatial super-resolution. In this paper, we attempt to design a spectral super-resolution network by taking advantage of two intrinsic properties of HSI. The first one is the spectral correlation. Based on this property, a decomposition subnetwork is designed to reconstruct HSI. The other one is the projection property, i.e., RGB image can be regarded as a three-dimensional projection of HSI. Inspired from it, a self-supervised subnetwork is constructed as a constraint to the decomposition subnetwork. These two subnetworks constitute our end-to-end super-resolution network. In order to test the effectiveness of it, we conduct experiments on three widely used HSI datasets (i.e., CAVE, NUS, and NTIRE2018). Experimental results show that our proposed network can achieve competitive reconstruction performance in comparison with several state-of-the-art networks. [ABSTRACT FROM AUTHOR]

A flat-field correction method is proposed for multiple measured hyperspectral microscopy imaging in this paper. As the most crucial preprocessing process in quantitative microscopic analysis, flat-field correction solves the uneven illumination caused by vignetting in microscopic images, and guarantees the precision of spatial and spectral information in hyperspectral microscopic imaging. In order to carry out flat-field correction and extract uneven illumination among groups of hyperspectral microscopic data containing hundreds of bands simultaneously, two properties of vignetting have been exploited: i) low-rank property is reflected by little information contained in vignetting; ii) local smoothness can be observed as a gradual change in brightness of vignetting, which is typically equivalent to the sparseness in spatial frequency domain. Combining the two properties above, a novel Low Rank Smooth Flat-field Correction (LRSFC) model modified from common orthogonal basis extraction is proposed, while an optimization is solved based on alternating direction multiplier method (ADMM), obtaining a unique flat-field term with low-rank and smooth properties. Qualitative and quantitative experimental assessments indicate that LRSFC does not add extra cell texture to the extracted flat-field term, whose performance appears prior to other state-of-the-art flat-field correction methods. [ABSTRACT FROM AUTHOR]

In 2015, two major conferences focused on hyperspectral remote sensing and imaging spectroscopy were held in Luxembourg and Tokyo, respectively. They are namely the 9th EARSeL SIG Imaging Spectroscopy workshop and the 7th Workshop on Hyperspectral Image and Signal Processing. As a follow up to these conferences, the current special issue of JSTARS was launched and we are proud to present a selection of 48 excellent papers. The large number of papers, which were drawn from both extended versions of workshop presentations and traditional paper submissions, is indicative of the high level of research activity, interest, and applications in this field. The papers in this special issue cover a wide range of topics, including new imaging spectroscopy systems, research and applications of imaging spectroscopy, and advances in hyperspectral image and signal processing. [ABSTRACT FROM PUBLISHER]

Many papers in the hyperspectral literature use simulations (based on a linear mixture model) to test algorithms, which either estimate the “intrinsic” dimensionality (ID) of the data or endmembers. Usually, these simulations use “real-world” endmembers, proportions distributed according to a uniform or Dirichlet distribution on the endmember simplex, and Gaussian errors which are “spectrally” and “spatially” uncorrelated. When the error standard deviations (SDs) in different bands are assumed to be unequal, they are usually estimated using Roger's method. The simulated and real-world data in these papers are so different that one cannot be confident that the various advocated methods work well with real-world data. We propose a general methodology which produces more realistic simulations, providing us with greater insights into the strengths and weaknesses of various advocated methods. With the aid of the well-known Indian Pines and Cuprite scenes, we compare several specific options within the proposed methodological framework. We also compare the performance of five well-known ID estimators using both real and simulated datasets and demonstrate that Roger's SD estimates are positively biased. A proof that Roger's estimates are always positively biased is given. [ABSTRACT FROM PUBLISHER]

This paper presents a novel panoramic hyperspectral stereo imaging system with depth map estimation capability. Images are collected from two line scan hyperspectral cameras in stereo configuration, while the platform is rotating. The proposed system combines the strengths of hyperspectral, stereo, and panoramic imaging. In addition, the proposed system uses passive sensors and hence does not need any radiation source, unlike LiDAR-based systems. In this paper, the proposed system, which captures hyperspectral data from 400 to 1000 nm wavelengths, is introduced, the stereo calibration procedure is described, corresponding analytic analyses are carried out with relation to rotating line-scan camera theory, and critical system parameters and aspects are addressed. In addition, a novel multiband stereo matching approach, which introduces multiband Census transform and SAD-based cost aggregation to stereo matching literature, is proposed. Experimental results are evaluated using two hyperspectral datasets, and the performance of the proposed algorithm is compared with the performance of several prominent matching algorithms available in the literature. A LiDAR-based system is used to enable an analysis of depth information accuracy. The experimental results show the enhanced matching and depth estimation performance of the proposed matching approach over standard stereo methods, and the consistency of depth values between LiDAR and the proposed system. The proposed system, which combines the strengths of panoramic imaging and stereo imaging with the high spectral resolution of hyperspectral cameras, uses passive imaging solutions and is fit to be used in applications requiring depth computation, target detection, and change detection. [ABSTRACT FROM PUBLISHER]

This paper proposes an approach for the development of automatic object-based techniques used for hyperspectral image classification. The proposed approach employs an adaptive smoothing step that utilizes an extension of partial differential equations (PDEs) from real domain (RPDE) to complex domain (CPDE). This idea results in generalized PDEs that simultaneously have the properties of both forward and backward diffusions. The genetic algorithm and an innovative fitness function are applied for adaptively tuning the CPDE parameters. The smoothed data are then fed into a Pixon-based object extraction process, which is itself an adaptive process. We also propose a novel distance metric for the Pixon creation step in order to facilitate the use of textural information which exists in the data. The spectral features of the extracted objects are used for a support vector machine classifier. [ABSTRACT FROM AUTHOR]

Recent classification-oriented proposals to thematic maps building from hyperspectral images have used both semisupervised approaches and spatial information for correction of spectral classification. Semisupervised approaches enrich the training data set adding similar samples to each class, whereas spatial correction is based on the natural assumption of thematic class spatial compactness. In this paper, we propose and validate the following innovations: 1) a new spectral classifier, which is called anticipative hybrid extreme rotation forest (AHERF); 2) a spatial–spectral semisupervised approach; and 3) a final spatial classification correction step. The novel heterogeneous ensemble learning approach AHERF starts with a model selection phase, using a small subsample of the training data, in order to define a ranking-based selection probability distribution of the classifier architectures that will be used in the ensemble, so that the architecture best adapted to the data domain will be used more frequently to train individual classifiers in the ensemble. After this initial phase, AHERF trains a heterogeneous ensemble applying random rotations to bootstrapped samples of the remaining training data, aiming to obtain diversified and data-domain adapted individual classifiers. The natural assumption that spatially close pixels will most likely have highly correlated values is exploited in two phases of the process pipeline. First, semisupervised label assignment is supported by spectral similarity and spatial proximity. Unsupervised spectral similarity is detected by latent class discovery. In this paper, we use a clustering algorithm (i.e., k-means). Second, maximizing class spatial compactness removes classification errors that appear as speckle noise in the classification image. The whole approach aims to use minimal sets of labeled pixels for training, which we call the seed training data set. Testing results are computed over the entire image ground truth. For comparison, we provide results in several steps: 1) of classification by AHERF and competing classifiers built by semisupervised training and 2) after spatial correction. We validate the approach on several conventional benchmarking images, achieving results which are comparable with state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

In hyperspectral unmixing, bilinear and linear–quadratic models have become popular recently, and also the polynomial postnonlinear model shows promising results. These models do not consider endmember interactions involving more than two endmembers, although such interactions might compose a nontrivial part of the observed spectrum in scenarios involving bright materials and complex geometrical structures, such as vegetation and intimate mixtures. In this paper, we present an extension of these models to include an infinite number of interactions. Several technical problems, such as divergence of the resulting series, can be avoided by introducing an optical interaction probability, which becomes the only free parameter of the model in addition to the abundances. We present an unmixing strategy based on this multilinear mixing (MLM) model; present comparisons with the bilinear models and the Hapke model for intimate mixing; and show that, in several scenarios, the MLM model obtains superior results. [ABSTRACT FROM AUTHOR]

In this paper, we consider signal subspace estimation based on low-rank representation for hyperspectral imagery. It is often assumed that major signal sources occupy a low-rank subspace. Due to the mixed nature of hyperspectral remote sensing data, the underlying data structure may include multiple subspaces instead of a single subspace. Therefore, in this paper, we propose the use of low-rank subspace representation to estimate the number of subspaces in hyperspectral imagery. In particular, we develop simple estimation approaches without user-defined parameters because these parameters can be fixed as constants. Both real data experiments and computer simulations demonstrate excellent performance of the proposed approaches over those currently in the literature. [ABSTRACT FROM PUBLISHER]

Several methods based on Total Variation (TV) have been proposed for Hyperspectral Image (HSI) denoising. However, the TV terms of these methods just use various l1 norms and penalize image gradient magnitudes, having a negative influence on the preprocessing of HSI denoising and further HSI classification task. In this paper, a novel l0 Total Variation (l0 TV) is first introduced and analyzed for the HSI noise removal framework to preserve more information for classification. We propose a novel Tensor low-rank constraint and $l_0$ Total Variation (TLR-l0 TV) model in this paper. l0 TV directly controls the number of non-zero gradients and focuses on recovering the sharp image edges. The spectral-spatial information among all bands is exploited uniformly for removing mixed noise, which facilitates the subsequent classification after denoising. Including the Weighted Sum of Weighted Nuclear Norm (WSWNN) and the Weighted Sum of Weighted Tensor Nuclear Norm (WSWTNN), we propose two TLR-l0 TV-based algorithms, namely WSWNN-l0 TV and WSWTNN-l0 TV. The Alternating Direction Method of Multipliers (ADMM) and the Augmented Lagrange Multiplier (ALM) are employed to solve the l0 TV model and TLR-l0 TV model, respectively. In both simulated and real data, the proposed models achieve superior performances in mixed noise removal of HSI. Especially, HSI classification accuracy is improved more effectively after denoising by the proposed TLR- l0 TV method. [ABSTRACT FROM AUTHOR]

Hyperspectral imaging features an important issue in remote sens ing and applications. Requirement to collect high volumes of hyper spectral data in remote sensing algorithms poses a compression prob lem. To this end, many techniques or algorithms have been develop ed and continues to be improved in scientific literature. In this paper, we propose a recently developed lossy compression method whi ch is called tridiagonal folded matrix enhanced multivariance prod ucts representation (TFEMPR). This is a specific multidimensional array decomposition method using a new mathematical concept called “folded matrix” and provides binary decomposi tion for multidimensional arrays. Beside the method a comparati ve analysis of compression algorithms is presented in this paper by means of compression performances. Compression performance of TFEMPR is compared with the state-art-methods such as compressive -projection principal component analysis, matching pursu it and block compressed sensing algorithms, etc., via average peak signal-to-noise ratio. Experiments with AVIRIS data set indicate a superior reconstructed image quality for the propo sed technique in comparison to state-of-the-art hyperspectral data compression methods. [ABSTRACT FROM AUTHOR]

Image decolorization is known to be an effective way in transferring a color image into a gray one while well preserving the major information of all three bands. In this paper, a simple yet effective hyperspectral image visualization framework based on decolorization, named decolorization based hyperspectral visualization, is proposed, which enables us to fully exploit the benefits of decolorization technique. The proposed framework consists of the following two main steps. First, the hyperspectral image is partitioned into nine subsets of adjacent hyperspectral bands and the averaged band of each subset is calculated. Then, the dimension reduced image is further divided into three groups of adjacent bands, and the bands in each group are fused by using an image decolorization method. The main contribution of this paper is that the strong correlations in two different fields, i.e., image decolorization and hyperspectral image visualization, are first built. Experiments performed on several real hyperspectral data sets demonstrate that the proposed framework can obtain outstanding visualization performance in terms of both subjective and objective evaluations. [ABSTRACT FROM AUTHOR]

Hyperspectral image (HSI) noise reduction is an active research topic in HSI processing due to its significance in improving the performance for object detection and classification. In this paper, we propose a joint spectral and spatial low-rank (LR) regularized method for HSI denoising, based on the assumption that the free-noise component in an observed signal can exist in latent low-dimensional structure while the noise component does not have this property. The proposed HSI denoising method not only considers the traditional LR property across the spectral domain but also leverages nonlocal LR property over the spatial domain. The main contribution of this paper is the incorporation of the low-rankness-based nonlocal similarity into sparse representation to characterize the spatial structure. Specially, the similar patches in each cluster usually contain similar sharp structure such as edges and textures; LR performed on cluster entitles to achieve a lower rank than that on the global spectral correlation. To make the proposed method more tractable and robust, we develop a variable splitting-based technique to solve the optimization problem. Experiment results on both simulated and real hyperspectral data sets demonstrate that the proposed method outperforms state-of-the-art methods with significant improvements both visually and quantitatively. [ABSTRACT FROM AUTHOR]

Changes in mangrove forests are mostly caused by natural disasters, anthropogenic forces, and uncontrolled population growth. These phenomena often lead to a strong competition between existing mangrove species for their survival. Due to its enhanced spectral resolution, remotely sensed hyperspectral image data can play an important role in the task of mapping and monitoring changes in mangrove forests. In this paper, we use EO-1 Hyperion hyperspectral data to model the abundance of pure and mixed mangrove species at subpixel levels. This information is then used to analyze the interspecies competition in a study area over a period of time. Different characteristics, including the rate of growth, rate of reproduction, mortality rate, and other processes that control the coexistence of plant species are discussed. The obtained results, verified through field visits, illustrate that the proposed approach can interpret the dominance of certain mangrove species and provide insights about their state of equilibrium or disequilibrium over a fixed time frame. [ABSTRACT FROM PUBLISHER]

Hyperspectral imagery (HSI) exploitation typically requires spectral signatures for target detection and identification algorithms. As the longwave infrared (LWIR) region of the electromagnetic spectrum is dominated by thermal emission, spectral radiance measurements are influenced by object temperature, and thus, estimates of target temperature may be necessary for emissivity retrieval to support these algorithms. Therefore, lack of accurate temperature information poses a significant challenge for HSI target detection and identification. Previous studies have demonstrated LWIR hyperspectral unmixing in both radiance and emissivity domains using in-scene target signatures. Here, a radiance-domain LWIR material identification algorithm for subpixel target identification of solid materials is developed by combining spectral radiance and linear mixing models with Bayesian model averaging. Application to experimental LWIR HSI illustrates that the algorithm effectively distinguishes between solid materials with a high degree of spectral similarity and reduces the probability of false alarms by at least one order of magnitude over a standard adaptive coherence estimator detector. Limits of identification are inferred from the imagery and found to depend on material type, target size, and target geometry. For the sensor and materials in this paper, the results imply that targets of nominally 5 m2 in size with strong spectral features can be identified for ground sampling distances (GSDs) on the order of 5-10 m (with abundances as low as ~10%) whereas blackbody-like materials are difficult to distinguish for GSDs larger than approximately 3 m. [ABSTRACT FROM PUBLISHER]

Presents corrections to the article “Anomaly Detection in Hyperspectral Images via Regularization by Denoising”. [ABSTRACT FROM AUTHOR]

Endmember extraction (EE) is an important process in hyperspectral image processing, and swarm intelligence algorithms have been developed to provide effective solutions for EE. However, these algorithms have limitations in terms of their calculation efficiency. To increase the computing speed of these algorithms by fully utilizing their potential parallel nature, this paper uses the artificial bee colony (ABC) algorithm to develop a multiagent system (MAS) for extracting endmembers from hyperspectral images. In this paper, EE is described as an optimization problem that involves the simplex volume and root-mean-square error (RMSE), and the ABC algorithm is used to obtain the optimal solution to the problem. To accelerate the execution of the ABC algorithm, it is incorporated into an established MAS platform that provides the advantages of high parallel computing efficiency, flexible system architecture, and responsive fault tolerance. Artificial bees and food sources, which are the two key components of the ABC algorithm, are implemented as standalone software agents. Different agents cooperate with each other via communication and produce the optimal solution. Comparative experiments are conducted to evaluate the performance of the agent-based ABC approach. The results indicate that the proposed agent-based ABC method can effectively solve the EE problem in distributive and high-speed computing environments. [ABSTRACT FROM PUBLISHER]

Noise contamination is a ubiquitous problem in hyperspectral images (HSIs), which is a challenging and promising theme in many remote sensing applications. A large number of methods have been proposed to remove noise. Unfortunately, most denoising methods fail to take full advantages of the high spectral correlation and to simultaneously consider the specific noise distributions in HSIs. Recently, a spectral–spatial adaptive hyperspectral total variation (SSAHTV) was proposed and obtained promising results. However, the SSAHTV model is insensitive to the image details, which makes the edges blur. To overcome all of these drawbacks, a spectral–spatial kernel method for HSI denoising is proposed in this paper. The proposed method is inspired by the observation that the spectral–spatial information is highly redundant in HSIs, which is sufficient to estimate the clear images. In this paper, a spectral–spatial kernel regularization is proposed to maintain the spectral correlations in spectral dimension and to match the original structure between two spatial dimensions. Moreover, an adaptive mechanism is developed to balance the fidelity term according to different noise distributions in each band. Therefore, it cannot only suppress noise in the high-noise band but also preserve information in the low-noise band. The reliability of the proposed method in removing noise is experimentally proved on both simulated data and real data. [ABSTRACT FROM PUBLISHER]

Hyperspectral data classification is a hot topic in remote sensing community. In recent years, significant effort has been focused on this issue. However, most of the methods extract the features of original data in a shallow manner. In this paper, we introduce a deep learning approach into hyperspectral image classification. A new feature extraction (FE) and image classification framework are proposed for hyperspectral data analysis based on deep belief network (DBN). First, we verify the eligibility of restricted Boltzmann machine (RBM) and DBN by the following spectral information-based classification. Then, we propose a novel deep architecture, which combines the spectral–spatial FE and classification together to get high classification accuracy. The framework is a hybrid of principal component analysis (PCA), hierarchical learning-based FE, and logistic regression (LR). Experimental results with hyperspectral data indicate that the classifier provide competitive solution with the state-of-the-art methods. In addition, this paper reveals that deep learning system has huge potential for hyperspectral data classification. [ABSTRACT FROM PUBLISHER]