The low-rank and sparse matrix decomposition (LSMD) approach effectively separates background components from anomalies in hyperspectral anomaly detection (HAD), suppressing background interference and enhancing target saliency. However, the high dimensionality, spectral redundancy, and massive data volume pose significant challenges for processing hyperspectral images (HSI). To address these issues, selecting efficient dimensionality reduction strategies for concise and effective extraction of background and anomaly information is crucial for robust anomaly detection. In this paper, we use compressed sensing technology to reduce the dimensionality of HSI data and establish a go decomposition-based iterative multiple random compressed sampling anomaly detection method (IMRCSGD). The IMRCSGD applies the low-rank and sparse matrix decomposition in compressed sampling band domains, and performs RX and RAD anomaly detection in a compressed composite subspace. Specifically, IMRCSGD employs multiple compressive sampling processes to preserve as much significant spectral-spatial information as possible. In addition, a novel iteration mechanism is introduced to reduce the random characteristic and improve the detection performance by incorporating feedback from the texture and content of the newly generated detection results. The effectiveness of the IMRCSGD is validated by experimental results obtained from six real hyperspectral datasets.

Due to the limitations of satellite imaging hardware, it is more demand to fuse hyperspectral image (HSI) and multispectral image (MSI) in an unsupervised manner in real application scenarios. However, current research has certain limitations, either relying on more prior knowledge or lack of verification of model performance with real satellite data. In order to solve the above problems, this paper designs a deep unsupervised adaptive attention network (DUAA-Net). This model can adaptively learn the mapping relationship between HSI and MSI. Compared with related unsupervised deep learning methods, this model not only simulates data achieved better indicators but also more accurate results in classification applications of real data.

The low-rank and sparse decomposition has found widespread application in hyperspectral image (HSI). To further improve the anomaly detection effect, an accurate and appropriate decomposition strategy is necessary. In this paper, we develop a go decomposition-based model with independent component analysis (ICA) for hyperspectral anomaly detection, referred as ICGD. This model utilizes data sphering to remove the first two order statistics, such that better represent HSI. Then, inspired by go decomposition thought, the pure low-rank background component and sparse anomaly component are generated. Through combining the obtained low-rank item and sparse item as the input of RX, anomaly detection can be realized. And experimental result on real hyperspectral data reveals the effectiveness of ICGD.

Recently, the methods based on self-attention mechanisms have gained increasing prominence in hyperspectral image classification (HSIC). However, the existing self-attention mechanism suffers the challenge of attention shift and redundancy. To address the problem, we propose the center category focusing transformer network (CCSF-Transformer) for HSIC, which is designed to resolve attention shifts and redundancy by balancing the multiple category features. Specifically, the central-category-focused attention mechanism (CFA) is presented in the proposed framework to compute the category-matched attention between the center pixel and neighbor pixels, closely matching the center-pixel style labeling strategy, and reducing the computation complexity by excluding the computation between interference pixels. Besides, the spectral-salient-focused attention module (SFA) is developed to capture the spectral correlation, which concentrates on the salient bands and suppresses the expression of redundant bands. Moreover, the hierarchical integration network (HIN) is built to rectify spatial and spectral features The experiment results on two popular HSI datasets demonstrate that the proposed method achieves robust performance compared to other state-of-the-art methods.

Change detection (CD) in remote sensing imagery is identified as a pivotal task in the field of Earth observation, while it usually confronts the dilemma of intricate data and minor alterations. To address the stated challenge, this letter presents an innovative frequency-temporal attention network for CD (FTAN), which incorporates two advanced modules including the multidimensional convolutional frequency attention module (MCFA) and the interactive attention module (IAM). Specifically, the MCFA module is essential for enhancing sensitivity in CD by merging multiscale spatial and frequency domain features. As a supplement to MCFA, the IAM aggregates category-related tokens and processes cross-attention information from different time phases. The seamless integration of MCFA and IAM empowers the FTAN network with enhanced capabilities to detect minor regions and edges accurately. Experiments on datasets like LEVIR-CD and DSIFN-CD demonstrate superior performance by outperforming existing models in F1 scores and IoU metrics. Our code and pretrained models will be released athttps://github.com/chirsycy/FTAN.

The sparsity constraint of abundance plays an important role in the effectiveness of nonnegative matrix decomposition for hyperspectral unmixing (HU).$L_{0}$norm,$L_{1}$norm, and$L_{2}$norm are commonly used as sparsity regulation items individually or in combination. But they suffer from the problems of NP-hard optimization, uncontinuous differentiability, or poor sparsity, degrading the unmixing accuracy. Choosing an appropriate$L_{p}$norm can promote sparse unmixing of hyperspectral data. However, determining the value of p and speeding up its optimization process is essential. This letter proposes a multilayer nonnegative matrix factorization method for HU using chaotic sequences (CMLNMF) to optimize the$L_{p}$norm, which has strong randomness and good traversal. Introducing chaotic sequences makes it possible to induce the sparsity of$L_{p}$norm in unmixing quickly. On the simulated dataset, the capability of chaotic sequence on optimizing$L_{p}$norm is compared with that of conventional uniform distribution. The experiment shows that the chaotic sequence obtains the optimal solution through two searches, with higher efficiency than a uniform distribution. The algorithm is compared with state-of-the-art unmixing methods on simulated and real datasets as a whole. It quantitatively demonstrates the superiority and effectiveness of CMLNMF over several methods.

Blurring and noise degrade the performance of image processing. To mitigate this effect, various regularization-based deblurring methods have been proposed. Total variation regularization is widely used owing to its excellent ability in preserving the salient edges, but it also tends to smooth the image details. In this paper, we propose a local extremum-constrained total variation (LECTV) framework for image deblurring. In the developed deblurring framework, we integrate prior knowledge of the dark channel with the structural features of the image into a single regularization term. Furthermore, unlike most existing methods that focus on the overall sparsity of the dark channel, the defined regularization term allows for a pixel-wise adaptive description of the image to restore its inherent spatial texture structure. Finally, a majorization-minimization-based method is designed to solve the developed LECTV framework. Experimental results on natural and hyperspectral images show that the designed framework exhibits excellent performance in removing multiple types and degrees of blurring. Extensive evaluations also further show its superiority compared to other advanced methods.

For scenes with complex backgrounds and weak anomalies, how to effectively distinguish anomaly targets from the background is the key to perform hyperspectral image anomaly detection (AD). Data decomposition-based methods have been widely studied due to their potential in separating background and anomaly components. However, due to its unclean background extraction and sensitivity to noise, it has an adverse effect on the detection of anomaly targets. Additionally, a large amount of spectral data can lead to an increase in computation during data decomposition. To address this issue, we propose an AD method based on a feedback band group and variation low-rank sparse model (FBGVLRS-AD). Firstly, we employ a uniform band selection strategy to partition spectral bands and perform data decomposition on the selected band group, to separate low-rank and sparse components. This decomposition on the band group can reduce computational time and mitigate the interference from spectral variability. Secondly, to preserve the integrity of abnormal target spectra during the background extraction process, theL2,1norm is employed for joint correlated total variation to extract the desired anomalous targets. Then, utilizing the detection information from the existing band groups, a feedback-driven iterative framework has been designed to consider the consistency and complementarity in AD across band groups. This framework facilitates the extraction of sparse components in subsequent band groups and reinforces the anomalous elements. Iteratively addressing these sub-problems on band groups helps prevent the loss of useful spectral information, maintaining sufficient anomaly information while reducing interference from redundant information and spectral variations. Finally, the proposed FBGVLR-AD is optimally solved by the augmented Lagrange multiplier (ALM) method. Comparison with state-of-the-art anomaly detectors on multiple data validates the competitiveness of the proposed method for AD tasks.

Real-time anomaly detection technique can efficiently and effectively leverage available data and operates in tandem with data collection, avoiding dependence on unacquired spectral data. Nonetheless, there are no restrictions or discussions regarding the scope of utilization for existing data. Overloading the analysis with excessive information, particularly encompassing dynamically changing background scenes, can introduce interference, undermining the statistical characteristics of the data and hampering anomaly detection. Studies indicate that local anomaly detection can enhance detection performance. Consequently, determining the optimal scope of the row space within the context of real-time line-by-line processing by integrating local processing and real-time technology stands pivotal in enhancing efficacy. In order to realize real-time hyperspectral local anomaly detection, based on the most widely used push-broom hyperspectral imaging sensor, this article proposes a finite Markov local real-time correlation matrix$R$anomaly detection (FMLRT-RAD) by studying the similarity of spectra in adjacent regions of the same substance and the independence of spectra in different regions in hyperspectral images. FMLRT-RAD can adaptively determine the size of the local background region, and solve the challenging task of selecting a suitable data range for local background suppression when an imaging sensor obtains a large amount of data. In addition, based on sample correlation matrix$R$anomaly detection, two different correlation matrix representations are designed for dynamically updating finite local samples. Woodbury matrix identity is used to update background suppression, and corresponding update equations with causal recursion characteristics are derived to achieve local real-time anomaly detection to further reduce time consumption and improve detection capability. The experimental results of several real-world hyperspectral image datasets show that the detector has superior detection performance compared with other advanced detectors.

Oriented to adaptive recognition of the new land-cover categories, incremental classification (IC) that aims to complete adaptive classification with continuous learning is urgent and crucial for hyperspectral image classification (HSIC). Nevertheless, deep-learning-based HSIC models adopted the learning paradigm with fixed classes yield unsatisfactory inference in the situation of IC due to the catastrophic forgetting problem. To eliminate the recognition gap and maintain the old knowledge during IC, in this paper, we propose a novel approach called the distillation-constrained prototype representation network (DCPRN) for hyperspectral image incremental classification (HSIIC). The primary goal of DCPRN is to enhance the discriminative capability for recognizing the original classes in HSIIC, while effectively integrating both the original and incremental knowledge to facilitate adaptive learning. Specifically, the proposed framework incorporates a prototype representation mechanism, which serves as a bridge for knowledge transfer and integration between the initial and incremental learning phases of HSIIC. Additionally, we present a dual knowledge distillation module in incremental learning, which integrates discriminative information at both the feature and decision level. In this way, the proposed mechanism enables flexible and dynamic adaptation to new classes and overcomes the limitations of fixed-category feature learning. Extensive experimental analysis conducted on three popular data sets validates the superiority of the proposed DCPRN method compared with other typical HSIIC approaches.

Hyperspectral image classification (HSIC) models have made remarkable progress in the last decade. Nevertheless, the downsized mapping in the convolutional neural network (CNN) and down-sampled mechanism in the transformer-based approach amplify the loss of hidden knowledge in the subpixel that encompasses crucial yet unseen features within a single pixel. Considering this aspect, the mentioned popular solutions for HSIC contradict the inherent characteristic of hyperspectral data. To address this issue, we rethink the size factor in CNN and propose a novel spatial mapping expansion with spectral compression (SMESC) network for HSIC. Specifically, the SMESC builds a mapping expansion network to mine unseen information in subpixels with enlarged feature maps. A channel modulation residual block (CMRB) is developed to compress spectral redundancy and promote salient channels with modulation information. Moreover, we design a multiple-size training strategy to substitute the traditional multiple feature extraction (FE) branches and improve the model adaptation to the different sizes of the testing samples. The extensive experimental results and analysis of four hyperspectral image (HSI) datasets demonstrate the superiority of the proposed architecture compared to other advanced HSIC methods. Our code will be released athttps://github.com/Chirsycy/SMESC.

Hyperspectral image (HSI) denoising is a crucial step for its subsequent applications. In this article, we propose TDSAT, a 3-D spatial-spectral attention Transformer model designed to effectively remove noise in HSI processing while preserving essential spectral and spatial information. The primary objective of this model is to utilize the 3-D Transformer to explore the global spectral-spatial features in HSI, learn the relationships among different bands, and preserve high-quality spectral and spatial information for denoising. The proposed method consists of three main components: the multihead spectral attention (MHSA) module, the gated-dconv feedforward network (GDFN) module, and the spectral enhancement (SpeE) module. The MHSA module learns the relationships among different bands and emphasizes the local spatial information. The GDFN module explores more expressive and discriminative spectral features. The SpeE module enhances the perception of subtle differences between different spectrums. Moreover, unlike the previous Transformer denoising method that can only handle fixed bands, the proposed method combines 3-D convolution and spectral-spatial attention Transformer blocks, enabling the denoising of HSI with an arbitrary number of bands. Experimental results demonstrate that TDSAT outperforms compared methods. The code is available athttps://github.com/Featherrain/TDSAT.

Mixed noise pollution in HSI severely disturbs subsequent interpretations and applications. In this technical review, we first give the noise analysis in different noisy HSIs and conclude crucial points for programming HSI denoising algorithms. Then, a general HSI restoration model is formulated for optimization. Later, we comprehensively review existing HSI denoising methods, from model-driven strategy (nonlocal mean, total variation, sparse representation, low-rank matrix approximation, and low-rank tensor factorization), data-driven strategy [2-D convolutional neural network (CNN), 3-D CNN, hybrid, and unsupervised networks], to model-data-driven strategy. The advantages and disadvantages of each strategy for HSI denoising are summarized and contrasted. Behind this, we present an evaluation of the HSI denoising methods for various noisy HSIs in simulated and real experiments. The classification results of denoised HSIs and execution efficiency are depicted through these HSI denoising methods. Finally, prospects of future HSI denoising methods are listed in this technical review to guide the ongoing road for HSI denoising. The HSI denoising dataset could be found at https://qzhang95.github.io.

Hyperspectral images provide spectral curves that reflect the "fingerprint" properties of substances, making them suitable for many applications. Thanks to the rapid development of computing resources, deep learning algorithms can significantly improve the cognitive ability of the network by extracting hidden features, and have been successfully applied to hyperspectral image processing, such as classification and detection. In this paper, a new hyperspectral target detection model based on one-dimensional generative adversarial networks (1D-GAN) is proposed. The proposed 1D-GAN network is designed to extract HSI features, and the probability is calculated accordingly whether the pixel to be detected is a target or background. In order to capture the spatial features, the guided filter is then used to obtain the final detection map. Performance comparison with several state-of-the-art methods demonstrate the effectiveness and efficiency of the proposed 1D-GAN algorithm.

Hyperspectral image (HSI) has attracted much attention because of its rich spectral information. However, due to the limitation of imaging hardware conditions, it is often difficult to directly obtain a high spatial resolution hyperspectral image (HR-HSI). To improve the resolution, it is an economical and effective method to fuse the hyperspectral image with the high spatial resolution multispectral image (HR-MSI) collected from the same scene. In recent years, with the development of deep learning, the convolutional neural network (CNN) based models have been applied to solve the super-resolution reconstruction of hyperspectral images. However, limited by the convolution kernel size, the receptive field of CNN is relatively small with more attention to the local information of the image. In order to solve this problem, this paper proposes a Swin Transformer based super-resolution reconstruction (STSR) network for hyperspectral images. Specifically, Swin Transformer structure is innovatively used in STSR as the skeleton of the network, where the Swin Transformer residuals are used to extract the global spatial feature information in the image. In addition, in order to retain the spectral details in the process of super-resolution reconstruction, a spectral attention module is introduced to preserve the original spectral information. The experimental results show that the high-resolution hyperspectral images fused by the proposed STSR method are superior to the comparison method in terms of vision and quality, which proves the superiority of this method.

Multi-exposure image fusion (MEF) technique is the most widely used method to obtain high dynamic range (HDR) images. Inspired by the recent successful application of Transformer in image processing, a hybrid dense connection network based on CNN and Transformer is proposed for MEF in this paper. Considering the importance of texture details to the multi-exposure image fusion task, shallow features containing rich texture details is also added to each dense layer, which are extracted by the pre-trained RepVGG. In addition, the dynamic weight calculation module is improved, so that different source images can obtain finer weight in the calculation of the loss function. Experiments are conducted on the dataset provided by MEFB, and both qualitative and quantitative comparisons show that the proposed method can achieve better results compared with the state-of-the-art algorithms.

Mixed noise pollution severely disturbs hyperspectral image (HSI) processing and applications. Plenty of algorithms have been developed to address this issue via two strategies: model-driven or data-driven strategy. However, model-driven methods exist in the highly time-consuming weakness of iterative optimization and unstable sensitivity of setting parameters. Data-driven methods usually perform poor due to the overfitting effects. To solve these issues, we combine both the deep denoising priors with low-rank tensor factorization (DP-LRTF) for HSI restoration. The proposed method uses Tucker tensor factorization to depict the global spectral low-rank constraint. Then the spectral orthogonal basis and spatial reduced factor are optimized by two deep denoising priors, respectively. Through this integrated strategy, we can simultaneously exploit the intrinsic low-rank property of HSI, and utilize the powerful feature extraction ability by deep learning for HSI restoration. Compared with model-driven and data-driven methods, DP-LRTF outperforms on HSI mixed noise removal and execution efficiency for various simulated/real experiments.

Hyperspectral images (HSIs) are often disturbed by various kinds of noises. This paper proposes a global variation and local structure low-rank model (GLLR) for HSI denoising by integrating spatial segmentation smoothing and spectral low-rank properties. Compared with existing denoising methods, the proposed method considers not only the global low-rank property but also the local structure low-rank property of HSIs. Specifically, the GLLR describes the global correlation and segmental smoothing structure of the HSI by the correlated total variation. In addition, we construct a new structural low-rank prior, called the local minimum difference (LMD) low-rank. With LMD low-rank property of HSI, GLLR can remove noise while retaining useful structural information in the HSI. Then, an ALM-based optimization algorithm is devised to solve the objective functions for the presented model. Finally, comparison experiments with existing methods are conducted on synthetic and real datasets to demonstrate the effectiveness and superiority of the proposed method.

Anomaly detection, as one of the most critical tasks in hyperspectral image (HSI) processing, has been paid extensive attention in the past decades. The classical work of hyperspectral anomaly detection, sample correlation matrix R-based anomaly detection (R-AD), can achieve desirable detection accuracy, but its processing complexity remains a challenging problem. Real-time anomaly detection methods speed up the procedure through processing the data acquired by grating splitting or acousto-optic tunable filter (AOTF)-based imaging systems in the manner of pixel-by-pixel, line-by-line, or band-by-band. However, in practical industrial scenarios, the multiarray filter spectral imaging system is generally leveraged to avoid high construction and maintenance costs, which acquires multiline multiband (MLMB) data lacking a real-time processing version. In addition, the background of industrial assembly lines is chaotic, and the changes between batches are rapid, both increasing the difficulty of accurate anomaly detection online. To cope with these challenges, a real-time online processing version [real-time multiline multiband R anomaly detection (RTMLMB-RAD)] of R-AD based on MLMB data is proposed here for the first time. Specifically, a multiline multiband correlation matrix (MLMBCM) is designed to update the detector of R-AD recursively with MLMB data acquisition. Since MLMBCM relies only on the previous result and the current data, a large amount of data storage and complex matrix inversion calculation are avoided. At the same time, local anomaly detection mode is adopted to improve the sensitivity of anomalies in different batches and different types of products. The experimental results on six public hyperspectral datasets demonstrate that the proposed RTMLMB-RAD algorithm outperforms other state-of-the-art methods in terms of time cost and detection accuracy. Particularly, the proposed algorithm was tested on the tobacco primary processing line with a detection rate of 87% and a false alarm rate of 10% at a speed of 3 m/s, further verifying the algorithm’s feasibility and effectiveness in practical industrial scenarios.

The ubiquity of mixed pixels in hyperspectral images makes it difficult for traditional classification techniques to determine the spatial distribution of land cover classes accurately. Subpixel mapping (SPM) technology is an effective method to solve this problem. Aiming at taking the multiple scales and the spatial features into account, an SPM method based on multi-scale and multi-feature (MSMF) is proposed, so as to effectively improve the accuracy of SPM. Firstly, the maximum linearization index method of the non-redundant complete straight-line set is designed to identify the linear distribution feature of land-cover classes. And then, different methods are applied to different spatial features and unified together finally, where the template matching iterative exchange is used for the linear distribution classes, and the multiscale spatial dependence iterative exchange method combined with area perimeter is used for the planar distribution classes. Experiments on three remote sensing images are carried out to evaluate the performance of MSMF. The results show that the proposed method can effectively improve the accuracy of SPM.

Deep learning-based hyperspectral target detection (HTD) methods are limited by the lack of prior information. Self-supervised learning is a kind of unsupervised learning, which mainly mines its own self-supervised information from unlabeled data. By training the model with such constructed valid posterior information, a valuable representation model can be learned and can get rid of the dependence of deep models on prior information. To this end, this article proposes a self-supervised spectral-level contrastive learning-based HTD (SCLHTD) method to train a model with spectral difference discrimination capability for HTD in a self-supervised manner. First, the hyperspectral images (HSIs) to be detected are sampled in odd and even bands, and the obtained band subsets are then used to train the corresponding adversarial convolutional autoencoders. Feature extraction part of the trained encoder is then used as the data augmentation function, where the positive and negative pairs are constructed through data augmentation, and the backbone is used to extract the representative vectors of the augmented samples. Second, the representative vectors are mapped to the spectral contrast space using spectral contrastive head, where the similarity and dissimilarity of spectra are learned by maximizing the similarity of positive pairs while minimizing the similarity of negative pairs, so that the backbone can discriminate spectral differences. Finally, aiming at suppressing the background, edge-preserving filters are used in conjunction with space information to process the detection results acquired by utilizing spectrum information via cosine similarity to generate the final detection results. Experimental results illustrate that the proposed SCLHTD method can achieve superior performances for HTD.

Hyperspectral unmixing could provide decomposition for small units in hyperspectral image, allowing accurate analysis of ground objects. Unfortunately, interference such as noise and spectral variability prevalent in hyperspectral data poses a serious challenge for it. Accordingly, this paper proposes a spectral-spatial anti-interference nonnegative matrix factorization (NMF) algorithm (SSAINMF), which improves the performance of spectral unmixing from both spectral and spatial perspectives. Specifically, the original data is analyzed and transformed into a statistical domain where the information of each dimension can be re-expressed, followed by a proof of restricted isometric and restricted isospectral properties for endmembers and abundances between the original domain and the transformation domain. To obtain more reliable endmembers, weighting is then applied to each dimension in the transformation domain depending on the priority coefficients quantified by their contribution to data representation, with the influence of anomalous and noisy data weakened and the priorities of low-rank information emphasized. Finally, superpixels are exploited to induce local similarity and structural sparsity of abundances within the neighborhood, which reduces the sensitivity to spatial noise and spectral variability. From experimental results on synthetic and real data sets, the proposed SSAINMF has demonstrated effectiveness in decomposing mixed pixels, with better robustness.

With the development and progress of deep learning, the use of deep learning technology for hyperspectral target detection has achieved excellent results. However, most deep-learning-based methods do not effectively suppress background. This paper presents a contrastive learning-based hyperspectral target detection (CLHTD) for this purpose. The positive and negative pairs are constructed through data augmentation, and the backbone is used to extract the representative vectors of the augmented samples. Then the representative vectors are mapped to the spectral and the cluster contrast space using their corresponding contrastive head, respectively. In the contrast space, the similarity and dissimilarity of spectra and clusters are learned by maximizing the similarity of positive pairs while minimizing the similarity of negative pairs, to increase the difference between the representative vectors of target and background. Finally, the detection result is obtained through the cosine distance. Experimental results illustrate that the proposed CLHTD algorithm can achieve superior performances for hyperspectral target detection.

A novel multi-scale fusion maximum entropy subspace clustering (MFMESC) for hyperspectral image (HSI) band selection is proposed in this paper. Subspace clustering is combined as a self-expression layer with stacked convolutional autoencoder, so that subspace clustering working in linear subspaces can deal with complicated HSI data with nonlinear characteristics. Multiple fully-connected linear layers are inserted between the encoder layers and their corresponding decoder layers to promote learning more favorable representations for subspace clustering. A multi-scale fusion module is designed to guide the fusion of multi-scale information extracted from different layers to learn a more discriminative self-expression coefficient matrix. Furthermore, the maximum entropy regularization is introduced in the subspace clustering to promote the connectivity within each subspace. Experimental results demonstrate the superiority of the proposed model against state of-the-art methods.

Factors such as insufficient training samples, high-dimensional data features, and unbalanced data classes can degrade the accuracy of hyperspectral classification. To this end, this letter proposes an iterative training sample augmentation (ITSA) algorithm and a new classification model incorporating ITSA and maximum margin projection (ITSA-MMP). First, ITSA iteratively augments samples by a similar region clustering strategy (SRCS) integrating spatial-spectral metric. Then, box-plot for representative sample selection (BPRSS) is adopted to screen optimal samples for the final augmented sample set (ASS). Next, based on the ASS, MMP projects the hyperspectral image into a low-dimensional subspace to explore the local structure of the data manifold and improve the interclass separability of the data. Finally, the MMP-reduced data is classified by support vector machines. Experiments on two real hyperspectral datasets validate that ITSA-MMP can effectively increase the training sample set especially for small initial sample set and unbalanced dataset and obtain a higher classification accuracy.

This letter proposes an unsupervised band selection (BS) algorithm named residual driven BS (RDBS) to address the lack ofa prioriinformation about anomalies, obtain a band subset with high representation capability of anomalies, and finally improve the anomaly detection (AD). First, an anomaly and background modeling framework (ABMF) is developed via density peak clustering (DPC) to pre-determine the prior knowledge of the anomalies and background. Then, the DPC-based constraints are applied to R-Anomaly Detector (RAD), and three band prioritization (BP) criteria are derived to obtain the representative band subset for anomalies. Experiments on two datasets show the superiority of RDBS over other BS algorithms and verify that the obtained band subsets are strongly representative of anomalies.

Unsupervised domain adaptation (UDA) attempts to boost the performance on an unlabeled target domain by transferring knowledge from a labeled source domain. The previous models consider domain-level discrepancy while neglecting content-level distinction. To further decrease the distribution gap between different domains, this letter proposes a novel UDA approach with content-wise alignment for hyperspectral image classification (HSIC). We accomplish feature alignment with content-wise discrepancy reduction through an adversarial framework for the first time. Expressly, the core of the proposed content-wise scheme is integrated with a class-level and style-perceive-level regularized alignment to strengthen the representation of invariant feature. The experimental analysis demonstrates that the proposed model achieves more effective performance than other domain adaptation methods for hyperspectral image (HSI).

Band selection (BS) aims to choose a salient subset implied sufficient information from the numerous bands, which supplies a significantly efficient way to alleviate the barrier of dimensionality disaster for hyperspectral image classification (HSIC). This letter develops a semisupervised BS approach based on dual-constrained low-rank representation BS (DCLRR-BS) with two regularizations for HSIC. To be specific, a low-rank representation model is first proposed with super-pixel and imbalanced class-wise constraints, which are explicitly integrated to improve the performance of the band description. Next, the clusters are built adaptively based on graph theory in an unsupervised manner to rapid selection efficiency. A selection criterion is last designed to highlight the prominent band of each subset cluster to fulfill the BS procedure. Experimental results conducted on four types of classifiers with two real hyperspectral image (HSI) data sets demonstrate that the proposed DCLRR-BS method performs well in the imbalanced HSIC area.

Convolutional neural networks (CNN) have been widely used in hyperspectral image classification (HSIC). Although the current CNN-based methods have achieved good performance, they still face a series of challenges. For example, the receptive field is limited, information is lost in down-sampling layer, and a lot of computing resources are consumed for deep networks. To overcome these problems, we proposed a multi-level spectral-spatial transformer network (MSTNet) for HSIC. The structure of MSTNet is an image-based classification framework, which is efficient and straightforward. Based on this framework, we designed a self-attentive encoder. Firstly, HSIs are processed into sequences. Meanwhile, a learned positional embedding is added to integrate spatial information. Then, a pure transformer encoder is employed to learn feature representations. Finally, the multi-level features are processed by decoders to generate the classification results in the original image size. The experimental results based on three real hyperspectral data sets demonstrate the efficiency of the proposed method in comparison with the other related CNN-based methods.

Due to the enormous data and redundant information, how to process hyperspectral images reasonably and efficiently has become a research focus. This article proposes a sequential band fusion (SBF) approach for hyperspectral target detection and gives a detailed derivation of the fusion theory. Then four fusion algorithms—SBF based on band sequence (SBF-BSQ), SBF driven by an initial band (SBF-IBD), SBF based on band priority (SBF-BP), and SBF based on band selection (SBF-BS)—are introduced for application. Experimental results prove that the method proposed in this article can not only effectively improve the efficiency of target detection but also provide the trend of the detection value during the fusion process. Band fusion breaks through the limitation of band selection in the application and provides a new processing method for hyperspectral data.

To achieve real-time abundance estimation of hyperspectral images and improve the accuracy and efficiency of estimation, this article proposes a new band processing approach for abundance estimation, to be called sequential band fusion (SBF). To achieve SBF, a new band priority mechanism is proposed. It is derived from the concept of orthogonal subspace projection (OSP) by orthogonalizing undesired targets using projection, while minimizing the variance resulting from the background. By taking advantage of OSP, the interfering effects caused by all undesired targets can be eliminated and then the detector produced by a target of interest can be further used as a measure of prioritizing bands as well as a means of searching bands for this particular target. As a result, two ranking-based band priority criteria (RP), called MaxOSP-RP and MinOSP-RP, and two searching-based band priority criteria (SP), called sequential feed forward band search (SFBS) and sequential backward band search (SBBS), can be derived. We provide a detailed theoretical description and formula derivation of the SBF and combine it with band sequence (BSQ), RP and SP to propose three different fusion mechanisms, SBF-BSQ, SBF-RP and SBF-SP to make the fusion mechanism applied to different scenarios. Experimental results show that the proposed methods performs well for abundance estimation.

This article presents a new approach, called progressive band subset fusion (PBSF) for hyperspectral anomaly detection. Unlike band selection (BS) which selects bands according to band prioritization or band search strategies, PBSF fuses band subsets progressively during data collection processing. It is completely opposite to BS that must be done after data are acquired and then select bands by removing spectral redundancy as post-data processing. To accomplish PBSF, two versions of PBSF are derived: PBSF of the multiple-band subset (PBSF-MBS) and PBSF of uniform BS (PBSF-UBS). In particular, the fusion process takes place in an anomaly detector from a real-time processing perspective. Three approaches are developed to realize PBSF of two-band subsets simultaneously: PBSF-band sequential (PBSF-BSQ), PBSF-RT, and PBSF-zigzag. Extensive experiments demonstrate that PBSF has advantages over BS in many ways.

This article presents a multispatial filtering module cascaded system (MSFMCS) for hyperspectral image classification (HSIC), which can serve as a paradigm to improve spectral–spatial classification. It includes multiple spatial filtering modules (SFMs) that are cascaded to particularly capture spatial information from the classification maps generated from the preceding modules. As a result, any spectral classifier (SC) can be used as an input to an initial/input module (IM). Through MSFMCS, its classification performance keeps improving as more SFMs are processed. To terminate MSFMCS, an automatic stopping rule is particularly designed by support vector machine (SVM) which is used not only as a classifier but also as a decision-maker. So, once an SC cannot be further improved, MSFMCS is terminated. One major benefit resulting from MSFMCS is its framework which can implement any arbitrary SC as its initial classifier in IM. Another is its ability in capturing additional spatial classification information module by module as the process progresses. A third one is no weights connected between modules so that no training phase is required like a feedforward neural network. Finally, the number of modules used in MSFMCS can be automatically determined by its stopping rule not predetermined empirically. To illustrate full advantages of MSFMCS in HSIC, three types of heterogeneous classifiers, pure-pixel-based SVM, mixed-pixel-based constrained energy minimization (CEM), and feature-extraction-based classifier—orthogonal total variation component analysis (OTVCA)—are used for experiments to demonstrate how MSFMCS can improve their classification performance.

Band selection can reduce information redundancy and improve application efficiency of hyperspectral images, such as classification. Traditional band selection methods usually weight different objectives and combine them together in a single function, making it difficult to balance conflict between various criteria. Recently, multiobjective optimization band selection techniques have got a lot of attention, which evaluate bands from different aspects separately, and search for a solution well balancing all the objectives simultaneously. However, few algorithms are focused on target detection, and most of them use just two objectives, which are not sufficient enough to select optimal band subset for detection. To alleviate these problems, this paper proposes an algorithm named target-oriented multiobjective optimization of band selection (TOMOBS). Firstly, the multiobjective optimization framework with three objective functions is modeled, involving information, noise, and correlation of the bands respectively. Secondly, in order to optimize the proposed model, the noninferior solution advantage matrix is designed based on the swarm intelligence optimization method, which can provide accurate solutions and improve the descriptiveness of multiobjective optimization problems. Thirdly, target-oriented evaluation mechanism is developed to guide selecting final result from the Pareto front, especially designed for target detection. Experiments on real hyperspectral datasets show that this algorithm can provide a subset of bands with strong representational capability for target detection, and achieve impressing results compared to the state-of-the-art methods.

This article proposes a new approach to hyperspectral band processing for anomaly detection, to be called sequential band fusion (SBF), derived from the band sequential (BSQ) data acquisition format used by a hyperspectral imaging sensor which fuses one single band at a time with previously fused band subset sequentially. In order to realize SBF, four versions, SBF-BSQ, initial band driven SBF (IBD-SBF), band prioritization SBF (BP-SBF), and band selection SBF (BS-SBF), are developed. Furthermore, to validate the sequentially fused results by SBF identical to that produced by combining all joint bands together, its mathematical theory and derivations are also presented. Finally, the full utility of SBF in anomaly detection is demonstrated through extensive experiments.

This article presents a bi-endmember semi-nonnegative matrix factorization (Semi-NMF) algorithm based on low-rank and sparse matrix decomposition (LRSMD), referred to as BLSNMF, to resolve the issues of endmember variability and nonlinear mixing. Given the fact that the hyperspectral images contain a large amount of redundant information, compressing sensing (CS) techniques can generally be used to randomly sense the effective information in an observed image according to the effective approximation of the bi-endmember components. In this article, the proposed BLSNMF integrates low-rank and sparse spaces decomposed by go decomposition (GoDec) or orthogonal subspace projection-based go decomposition (OSP-GoDec) with Semi-NMF to suppress interference between different components so as to improve the unmixing performance via a simple linear mixed model. Specifically, the observed data space is first decomposed by GoDec or OSP-GoDec to approximate four different attribute components, CS-sampled double low-rank components, structured sparse component, and noise component. Second, from the CS-sampled double low-rank components, the inherent and new endmembers along with their abundances are evaluated via Semi-NMF, and then, the double low-rank components are redescribed using the estimated endmembers and abundances. Finally, the serious interference entries in the structured sparse component space are removed from the data to better learn other attribute components. The experimental results show that BLSNMF can eliminate the interference of new endmembers and sparse noise so as to better evaluate the endmembers and abundances and effectively improve the ability to interpret the spectral information.

When predicting data for which limited supervised information is available, hyperspectral target detection methods based on deep transfer learning expect that the network will not require considerable retraining to generalize to unfamiliar application contexts. Meta-learning is an effective and practical framework for solving this problem in deep learning. This article proposes a new meta-learning based hyperspectral target detection using Siamese network (MLSN). First, a deep residual convolution feature embedding module is designed to embed spectral vectors into the Euclidean feature space. Then, the triplet loss is used to learn the intraclass similarity and interclass dissimilarity between spectra in embedding feature space by using the known labeled source data on the designed three-channel Siamese network for meta-training. The learned meta-knowledge is updated with the prior target spectrum through a designed two-channel Siamese network to quickly adapt to the new detection task. It should be noted that the parameters and structure of the deep residual convolution embedding modules of each channel in the Siamese network are identical. Finally, the spatial information is combined, and the detection map of the two-channel Siamese network is processed by the guiding image filtering and morphological closing operation, and a final detection result is obtained. Based on the experimental analysis of six real hyperspectral image datasets, the proposed MLSN has shown its excellent comprehensive performance.

Despite continuing to progress in hyperspectral image classification (HSIC) based on deep learning, the classification accuracy is limited to furtherly improve in the absence of labeled samples. To address this issue, the metric-based prototypical networks for few-shot learning have enjoyed widespread popularity. However, the conventional prototypical networks are vulnerable to the selected examples and fail to accomplish representative predictions for the prototypes in complicated situations. In this paper, we propose a multi-view calibrated prototype-learning framework for few-shot HSIC, which consists of three rectified strategies from different views to improve the robustness of prototypes in the embedding space. Specifically, the calibrated aggregation network is the first presented to calibrate the representations with local patches aggregation for the enhancement of the prototypes. Moreover, to improve the compactness of the intraclass expression, the calibrated metric learning with regularization terms is designed to strengthen the discrimination of the prototypes. Furthermore, we calibrate the feature distribution of supervised samples by transferring statistical knowledge to eliminate the local bias in the test phase. The extensive experimental results and analysis of three hyperspectral image datasets demonstrate the superiority of the proposed architecture compared with other advanced methods.

Hyperspectral image classification (HSIC) methods based on convolutional neural network (CNN) continue to progress in recent years. However, high complexity, information redundancy, and inefficient description still are the main barriers to the current HSIC networks. To address the mentioned problems, we present a spatial-spectral dense CNN framework with a feedback attention mechanism called FADCNN for HSIC in this article. The proposed architecture assembles the spectral-spatial feature in a compact connection style to extract sufficient information independently with two separate dense CNN networks. Specifically, the feedback attention modules are developed for the first time to enhance the attention map with the semantic knowledge from the high-level layer of the dense model, and we strengthen the spatial attention module by considering multiscale spatial information. To further improve the computation efficiency and the discrimination of the feature representation, the band attention module is designed to emphasize the weight of the bands that participated in the classification training. Besides, the spatial-spectral features are integrated and mined intensely for better refinement in the feature mining network. The extensive experimental results on real hyperspectral images (HSI) demonstrate that the proposed FADCNN architecture has significant advantages compared with other state-of-the-art methods.

The current hyperspectral image classification (HSIC) model based on the convolutional neural network for feature extraction and softmax classifier has been prone to the barrier of label prediction with limited samples. Substituting for the enormously complicated work of terrain labeling, few-shot learning provides a popular option for HSIC with very few annotated samples. In this paper, we proposed a novel edge-inferring framework with the meta-learning paradigm for hyperspectral few-shot classification (HSFSC). In which, a graph neural network for similarity measurement is firstly presented to iteratively infer edge labels with the exploitation of instance-level similarity and the distribution-level similarity. Besides, in the meta-training stage, the pixel prediction model and patch prediction model based on edge inferring architecture are concretized jointly to improve the classification accuracy of the test samples. Expressly, at the meta-testing phase, the dynamic task-guided self-diagnosis strategy is developed for the first time to diagnose the samples separability of the current classification task, which is responsible for dynamically assigning the most reliable results based on the generated reliability grade of the sample. The extensive experimental results and analysis of three hyperspectral image datasets demonstrate the superiority of the proposed HSFSC architecture compared with other advanced methods.

Hyperspectral image (HSI) provided with a substantial number of correlated bands causes calculation consumption and an undesirable "dimension disaster" problem for the classification. Band selection (BS) is an effective measure to reduce the information redundancy with the physics spectrum preserved for HSI. Although the existing BS methods have achieved noticeable progress, the correlation between neighbor bands still needs to be mined deeply for an effective selection criterion. This paper proposes a BS approach to collecting the discriminative band subset for hyperspectral image classification (HSIC), which adopts the self-supervised learning paradigm to implement the BS by auxiliary spectrum rebuilding task. In specific, we utilized a Convolutional neural network (CNN) and Graph Convolutional Network (GCN) for the spectral-spatial feature extraction. Next, GCN and CNN are developed for the refinement of the band correlation sequentially. Afterward, the selected bands in terms of the acquired correlation are fed into the presented self-supervised spectrum rebuilding network for spectral reconstruction. Simultaneously, the proposed architecture completed the selection with the optimization of the band reconstruction by a defined loss function. In this way, we supply substitution for selection criterion and path searching through the end-to-end framework. The extensive experimental results and analysis demonstrated that the proposed hybrid architecture provided a competitive band subset for the classification, and the accuracies with different types of classifiers are more effective than the compared BS methods.

Model-driven methods and data-driven methods have been widely developed for hyperspectral image (HSI) denoising. However, there are pros and cons in both model-driven and data-driven methods. To address this issue, we develop a self-supervised HSI denoising method via integrating model-driven with data-driven strategy. The proposed framework simultaneously cooperates the spectral low-rankness prior and deep spatial prior (SLRP-DSP) for HSI self-supervised denoising. SLRP-DSP introduces the Tucker factorization via orthogonal basis and reduced factor, to capture the global spectral low-rankness prior in HSI. Besides, SLRP-DSP adopts a self-supervised way to learn the deep spatial prior. The proposed method doesn't need a large number of clean HSIs as the label samples. Through the self-supervised learning, SLRP-DSP can adaptively adjust the deep spatial prior from self-spatial information for reduced spatial factor denoising. An alternating iterative optimization framework is developed to exploit the internal low-rankness prior of third-order tensors and the spatial feature extraction capacity of convolutional neural network. Compared with both existing model-driven methods and data-driven methods, experimental results manifest that the proposed SLRP-DSP outperforms on mixed noise removal in different noisy HSIs.

This paper proposes a new hyperspectral anomaly detection method based on transferred deep learning and tensor decomposition. Firstly, since there is no labeled input data for training in anomaly detection, the detection model is obtained by training convolutional neural network with transferred learning. Then the model is decomposed to increase the number of convolution layers, that is, the depth of the network, so as to give more accurate results without over fitting. At the same time, the spatial information of the input data is extracted in order to make full use of the existing data for detection. Finally, combining the spectral and spatial information of the current pixel, the detection result is given. Experiments on two hyperspectral datasets show that the proposed algorithm has excellent performance.

A large number of spectral bands in hyperspectral data can help to identify ground objects but also bring additional computational burden. To address this issue, this letter proposes a new target-constrained band selection approach with particle swarm optimization (TCPSOBS) to select a more representational band subset with low redundancy for target detection. TCPSOBS obtains the local and global optimal values of the particle swarm in the current iteration process by calculating the fitness function of each particle derived by constrained energy minimization (CEM), and iteratively updates the particle swarm to find the particle with the optimal band subset index. Experiments prove that TCPSOBS can effectively improve the detection accuracy compared to other most advanced methods.

This paper proposes a method to improve the performance of the supervised classification from two aspects. Firstly, the global spatial and local spectral similarity is used to extend the labeled sample size (GLS). Secondly, extended subspace projection (ESP) which projects the original image to a lower-dimensional subspace is used to alleviate band redundancy. Finally, the two implements are combined with the sparse representation classifier (SRC) to optimize the hyperspectral image classification (HSIC). The proposed method is named GLSESP. Experimental results on real hyperspectral data set demonstrate the practicality and effectiveness of GLSESP for HSIC tasks.

Iterative constrained energy minimization (ICEM) has shown success in classification. However, a drawback suffered from ICEM is its requirement of complete ground truth to calculate class means. This letter develops an iterative selection of training samples to extend ICEM with two versions: iterative fixed training sampling constrained energy minimization (CEM) (IFTS-CEM) which uses a fixed training sample set throughout the entire iterative process and iterative random training sampling CEM (IRTS-CEM) which uses a random training sampling (RTS) at each iteration. The experimental results demonstrate that IRTS-CEM performs better than IFTS-CEM and also comparable to ICEM.

Sparse representation (SR)-based models have shown to be a powerful category of frameworks for hyperspectral image classification (HSIC). However, current residual-driven methods mainly focus on the sparsity of the coefficient, which is generally used in conjunction with the dictionary. In fact, the discriminant information hidden behind the value of sparse coefficient is not fully exploited. In this letter, we analyze the SR-based framework from the perspective of sparse coefficient, develop the participation degree (PD)-driven decision mechanism, and establish a concise model called constraint representation (CR). Based on CR, an improved version called adjacent CR (ACR) is further proposed, with consideration of spatial coherence via adjacent constraint. Experimental results using two real hyperspectral datasets verify the improvements of the proposed methods over the other related models and their spatial variants.

Low-rank and sparsity-matrix decomposition (LRaSMD) has received considerable interests lately. One of effective methods for LRaSMD is called go decomposition (GoDec), which finds low-rank and sparse matrices iteratively subject to the predetermined low-rank matrix order m and sparsity cardinality k. This article presents an orthogonal subspace-projection (OSP) version of GoDec to be called OSPGoDec, which implements GoDec in an iterative process by a sequence of OSPs to find desired low-rank and sparse matrices. In order to resolve the issues of empirically determining p = m + j and k, the well-known virtual dimensionality (VD) is used to estimate p in conjunction with the Kuybeda et al. developed minimax-singular value decomposition (MX-SVD) in the maximum orthogonal complement algorithm (MOCA) to estimate k. Consequently, LRaSMD can be realized by implementing OSP-GoDec using p and k determined by VD and MX-SVD, respectively. Its application to anomaly detection demonstrates that the proposed OSP-GoDec coupled with VD and MX-SVD performs very effectively and better than the commonly used LRaSMD-based anomaly detectors.

Wealthy spectral information provided by hyperspectral image (HSI) offers great benefits for many applications in hyperspectral data exploitation. However, processing such high-dimensional data volumes that may result in redundant bands due to its high interband correlation will be a challenge. For target detection and classification, this is particularly true since there may only need a relatively small number of bands that respond one particular target of interest well, while most of other bands do not. Band selection (BS) is a major dimensionality reduction technique to remove the redundant bands and selects a few bands to represent the entire image. However, how to eliminate the effect of uninteresting targets with similar spectra on detection of interesting targets is a severe issue arising in target detection for BS. This article develops a new approach called target-constrained interference-minimized BS (TCIMBS) which can be used to select band subset for specific target detection, while annihilating targets of no interest and suppressing interferers and background. Its idea is derived from target-constrained interference-minimized filter (TCIMF). By taking advantage of TCIMF, two band prioritization (BP) criteria called forward minimum variance BP (FMinV-BP) and backward maximum variance BP (BMaxV-BP) along with their three band search-based BS counterparts called sequential forward TCIMBS (SF-TCIMBS), sequential backward TCIMBS (SB-TCIMBS), and improved SB-TCIMBS (SB-TCIMBS*) are derived. The experimental results suggest that TCIMBS can improve the detection accuracy and also achieve better performance in comparison with several state-of-the-art methods.

Low rank and sparsity matrix decomposition (LRaSMD) has received considerable interest lately. One of effective methods is called go decomposition (GoDec) which finds low rank and sparse matrices iteratively subject to a predetermined low rank order, m and a sparsity cardinality, k, In order to resolve issue of the empirically determined m and k, the well-known virtual dimensionality (VD) and a minimax-singular value decomposition (MX-SVD) developed in maximum orthogonal complement algorithm (MOCA) are used for this purpose. The constrained energy minimization (CEM) is used for experiments to demonstrate that the GoDec with VD and MX-SVD performs very effectively.

Classification is a major task in hyperspectral image (HSI) processing. This paper develops an approach by taking advantage of low rank matrix derived from the low rank and sparse matrix decomposition (LRSMD) model which decomposes a hyperspectral data matrix X as X = L+S+n where L, S and n are referred to low rank, sparse and noise matrices respectively. The hyperspectral image classification (HSIC) is then performed on the low rank matrix L rather than the original data matrix X where the well-known go decomposition (GoDec) is used to produce such LRSMD model. To determine the two key parameters used in GoDec, the rank of L, m, and the cardinality of the sparse matrix, k the well-known virtual dimensionality (VD) and minimax-singular value decomposition (MX-SVD) methods are used for this purpose. Finally, to demonstrate advantages of using the low rank matrix L, support vector machine (SVM) and an edge-preserving filters (EPF)-based classifiers are implemented to evaluate classification performance.

This paper develops band fusion techniques to fuse hyperspectral data from a data communication and transmission perspective. It can provide progressive profiles of fusing different bands and improve the efficiency by data processing. Anomaly detection is investigated for its application to demonstrate its utility. The experimental results prove that the proposed band fusion methods can not only ensure the accuracy of the detection results, but also can improve the data processing efficiency.

Hyperspectral imagery provides wealthy spectral information to make it suitable for many applications. However, for specific applications, extracting suitable bands from high-dimensional data is a tedious and difficult task. In the past, many methods have been developed to perform band selection for specific tasks such as target detection. However, there is very little work to consider and deal with the effects of suspected interfering targets. In this paper, a new method for band selection, called target-constrained interference-minimized band selection (TCIMBS) is developed for specific target detection. It can select a band set with strong characterization capabilities for desired targets and good suppression for undesired targets and background (BKG). Experimental results demonstrate that TCIMBS can improve the detection performance, and also achieve better performances in comparison with several state-of-the-art methods.

Sparse representation (SR)-based models have been widely applied for hyperspectral image classification. However, the original residual-driven frameworks ignore the property of sparse coefficient to some extent, and their spatial variants suffer obstacles of optimization due to the strong constraint. In this paper, based on previous works on sparse coefficient and its spatial expansion, we put forward a novel classifier, called superpixel-level constraint representation (SPCR). In particular, constraint representation (CR) is first applied to interpret the process of SR from perspective of participation degree (PD). Then, a relaxed and adaptive spatial constraint via superpixel segmentation is imposed to transform the individual PD to local relative activity degree (RAD). The final classification is determined based on a concise RAD-driven mechanism. Experimental results on real data set demonstrate the efficiency of the proposed method.

This letter introduces a new approach to hyperspectral image classification (HSIC), called progressive band selection processing of hyperspectral image classification (PBSP-HSIC), which performs classification in multiple stages in the sense that each stage performs HSIC progressively according to a specifically selected band subset. Interestingly, such PBSP-HSIC offers a rare view of how different classes are classified in progressive stages, which has never been explored in the past. The experimental results also show that PBSP-HSIC performs better than HSIC using full bands.

Hyperspectral image classification (HSIC) faces three major challenging issues, which are generally overlooked. One is how to address the background (BKG) issue due to its unknown complexity. Another is how to deal with imbalanced classes since various classes have different levels of significance, particularly, small classes. A third one is fractional class membership assignment (FCMA) resulting from a soft-decision classifier. Unfortunately, the commonly used classification measures, overall accuracy (OA), average accuracy (AA), or kappa coefficient are generally not designed to cope with these issues. This article develops a 3-D receiver operating characteristic (3-D ROC) analysis from a detection point of view to explore how these three issues can be resolved for HSIC. Specifically, it first develops one-class classifier in BKG (OCCB), called constrained energy minimization (CEM), and multiclass classifier in BKG (MCCB), called linearly constrained minimum variance (LCMV) in conjunction with 3-D ROC analysis to address the BKG issue. Then, by considering a small class as a signal to be detected, its class accuracy can be interpreted as signal detection power/probability so that the 3-D ROC analysis can be used to address the imbalanced class issue. Finally, FCMA can be treated as a detector by converting a soft-decision classifier to a hard-decision classifier in such a manner that the 3-D ROC analysis is also readily applied. The experimental results demonstrate that 3-D ROC analysis provides a very useful evaluation tool to analyze the classification performance.

Spectral-spatial framework has been widely applied for hyperspectral image classification task. Some well-established models, such as group sparse representation (GSR), have gained a certain advance but still mainly focus on the usage of local spatial similarity and neglect the nonlocal spatial information. Recently, nonlocal self-similarity (NLSS) has been exploited to support the spatial coherence tasks. However, current NLSS-based methods are biased toward the direct use of nonlocal spatial information as a whole, while the underlying spectral information is not well exploited. In this article, we proposed a novel method to exploit local spectral similarity through nonlocal spatial similarity, with the integration of local spatial consistency in a single framework. Specifically, the proposed approach first exploits the NLSS by searching the nonoverlapped similar patches in defined scopes. Then, spectral similarity is determined locally within the found patches. After that, the found similar data and the original data are fused in a designed pattern. Finally, the GSR-based classifier (GSRC) is applied to process the fused data characterized by the manifold learning algorithm. The experimental results based on three real hyperspectral data sets demonstrate the efficiency of the proposed method, with improvements over the other related nonlocal or local similarity-based methods.

This paper presents a new band selection approach, called uniform band interval divided band selection (UBIDBS) which uniformly divides a band range into a finite number of band intervals from which a band can be selected from each band interval according to a custom designed band prioritization (BP) criterion. Two BP criteria are introduced. One is derived from orthogonal subspace projection (OSP). The other is based on correlation matrix R originated from constrained energy minimization (CEM). These two criteria allow users to identify a most significant band to be selected in each of band intervals. As a result, it avoids band decorrelation required by BP to remove adjacent high- correlated bands.

Background (BKG) is an integral part of an image and has significant effect and impact on hyperspectral image classification (HSIC). Unfortunately, how to address the BKG issue has not received much attention over the past years. This paper investigates this issue by developing a mixed pixel classifier, iterative constrained energy minimization (ICEM) and a posteriori classification measure, called precision (PR).

This paper presents a class information (CI)-based band selection (BS) approach to hyperspectral image classification (HSIC). It introduces a new concept from an information theory point of view, CI which can be used to determine an appropriate weight imposed on each class of interest. Specifically, two types of criteria, intraclass information criterion (IC) and interclass IC are derived as CI probabilities to measure CI that can be used to determine the number of training samples required to be selected for each class. With such CI-calculated probabilities, another new concept called class self-information (CSI) is also defined for each class that can be further used to define the class entropy (CE) so that CSI and CE can be used to determine the number of bands required for BS, nBS. In order to find desired nBS bands, two types of BS methods based on CSI and CE are custom-designed, called single class signature-constrained BS (SCSC-BS) which utilizes the constrained energy minimization (CEM) to constrain each individual class signature to select bands for a particular class according to its CSI-determined nBS and a multiple class signatures-constrained BS (MCSC-BS) which takes advantage of linearly constrained minimum variance (LCMV) to constrain all class signatures to select CE-determined nBS bands for all classes. These SCSC-BS and MCSC-BS selected bands are then used to perform classification and evaluated by CI-weighted classification measures by real image experiments. The results show that HSIC using judiciously selected partial bands as well as CI-weighted measures can improve HSIC with using full bands.

This paper develops a new approach to band selection for multiple-target detection, called constrained-target band selection (CTBS). Its idea is derived from the concept of constrained energy minimization (CEM) by constraining a target of interest, while minimizing the variance resulting from the background (BKG). By taking advantage of CEM, the variance produced by a target of interest can be further used as a measure of prioritizing bands as well as a means of selecting bands for this particular target. As a result, two CTBS-based band prioritization (BP) criteria, called minimal variance-based BP (MinV-BP) and maximal variance-based BP (MaxV-BP), and two CTBS-based BS methods, called sequential forward CTBS (SF-CTBS) and sequential backward CTBS (SB-CTBS), can be derived for multiple-target detection. Since the bands selected by CTBS vary with targets of interest used to constrain CEM, in order for CTBS to be applied to multiple targets, a new fusion technique, called band fusion selection (BFS), is further developed for CTBS to integrate bands selected by different targets so that CTBS can work for all targets. Unlike most BS methods for target detection which generally simultaneously select a fixed set of bands for all targets of interest, the ideas of constraining multiple-target detection and using BFS are novelty of this paper. Experimental results show that CTBS performs well for multiple-target detection.

In hyperspectral image classification (HSIC), background (BKG) is generally excluded from consideration due to the fact that obtaining complete knowledge of BKG is nearly impossible in reality. Unfortunately, BKG has significant impact on classification and band selection (BS). This paper investigates both issues and presents a novel approach called class signature-constrained BKG suppression (CSCBS) approach to BS for HSIC, where class signatures can be obtained either by a priori or a posteriori knowledge or training samples, and BKG suppression can be accomplished by taking the inverse of the sample correlation matrix R. Its idea takes advantage of the concept of the linearly constrained minimum variance (LCMV) developed from adaptive beamforming by constraining class signatures of interest while minimizing the effect caused by the unknown BKG so as to enhance the classification performance. There are two immediate applications of CSCBS. One is its application to HSIC, in which it becomes a CSCBS classifier. The other is its use of the LCMV-suppressed BKG as a measure to derive the band prioritization (BP) criteria and BS. Experimental results demonstrate that generally CSCBS does not need the full-band set for HSIC since a partial band subset selected by CSCBS-BP/BS can actually improve the classification results using full-band information.

Anomaly detection (AD) generally finds targets that are spectrally distinct from their surrounding neighborhoods but cannot discriminate its detected targets one from another. It cannot even perform classification because there is no prior knowledge about the data. This paper presents a new approach to AD, to be called a posteriori AD for unlabeled anomaly classification where a posteriori indicates that information obtained directly from processing data is used as new information for subsequent data processing. In particular, a posteriori AD uses a Gaussian filter to capture spatial correlation of detected anomalies as a posteriori information which is included as new information for further AD. In doing so, a posteriori AD develops an iterative version of AD, referred to as iterative anomaly detection (IAD), which implements AD by feeding back Gaussian-filtered AD maps in an iterative manner. It then uses an unsupervised target detection algorithm to identify spectrally distinct anomalies that can be used to specify particular anomaly classes. To terminate IAD, an automatic stopping rule is also derived. Finally, it uses identified distinct anomalies as desired target signatures to implement constrained energy minimization (CEM) to classify all detected anomalies into unlabeled classes. The experimental results show that a posteriori AD is indeed very effective in unlabeled anomaly classification.

This paper develops a new Neyman–Pearson detection approach, to be called band-specified virtual dimensionality (BSVD), to estimating the number of bands required by band selection (BS),$n_{\mathrm {BS}}$, as well as finding desired bands at the same time. Its idea is derived from target-specified virtual dimensionality (TSVD) where targets under hypotheses as signal sources in TSVD are replaced with bands as signal sources and the test statistics derived for a Neyman–Pearson detector (NPD) is signal-to-noise ratio (SNR) that is used to derive orthogonal subspace projection (OSP) approach for hyperspectral image classification and dimensionality reduction. Accordingly, the resulting virtual dimensionality is referred to as OSP-based BSVD. Several benefits resulting from BSVD cannot be offered by the traditional BS methods. One is its direct approach to dealing with$n_{\mathrm {BS}}$. Another is no-search strategy needed for finding optimal bands. Instead, it uses NPD to determine and rank desired bands for band prioritization. Most importantly, it determines$n_{\mathrm {BS}}$and finds desired bands simultaneously and progressively.

Anomaly detection (AD) is designed to find targets that are spectrally distinct from their surrounding neighborhood. Unfortunately, commonly used anomaly detectors generally do not take into account its surrounding spatial information. This paper derives an iterative version of anomaly detection, iterative anomaly detection (IAD) to address this issue. Its idea is to use a Gaussian filter to capture spatial information of the anomaly detection map and then feeds back the Gaussian filtered AD map to create a new data cube. The whole process is repeated over again in an iterative manner. When IAD is terminated anomaly representatives are identified and can be used as desired target signatures to implement constrain energy minimization (CEM) so as to classify all detected anomalies. Accordingly, IAD can be considered as anomaly classification.

Automatic target generation process (ATGP) has been widely used for unsupervised hyperspectral target detection. It implements a succession of orthogonal subspace projections (OSPs) to extract targets of interest without prior knowledge. This paper extends ATGP to a kernel version of ATGP, called kernel ATGP (KATGP) to further deal with linear non-separation problem. It introduces nonlinear kernels to map original data space into a higher dimensional feature space so that ATGP can effectively find.

This paper extends target-constrained interference minimized filter (TCIMF) to multiclass-constrained background suppression classifier (MCBSC) for hyperspectral image classification. In order to capture spatial contextual information MCBSC makes use of a Gaussian filter to feed back a Gaussian-filtered MCBSC map to create a new set of hyperspectral images for MCBSC to be re-implemented again in an iterative manner, referred to as iterative MCBSC (IMCBSC). Finally, it uses Otsu's method to perform hyperspectral image classification. As shown by experiments, MCBSC generally performs better than existing spectral-spatial hyperspectral image classification techniques in terms of several quantitative measures, such as classification rate, false classification rate, precision rate, accuracy rate in addition to overall accuracy (OA) rate.

This paper presents a new approach, called band subset selection (BSS)-based hyperspectral anomaly detection (AD), which selects multiple bands simultaneously as a band subset rather than selecting multiple bands one at a time as the tradition band selection (BS) does, referred to as sequential multiple BS (SQMBS). Its idea is to first use virtual dimensionality (VD) to determine the number of multiple bands, nBS needed to be selected as a band subset and then develop two iterative process, sequential BSS (SQ-BSS) algorithm and successive BSS (SC-BSS) algorithm to find an optimal band subset numerically among all possible nBS combinations out of the full band set. In order to terminate the search process the averaged least-squares error (ALSE) and 3-D receiver operating characteristic (3D ROC) curves are used as stopping criteria to evaluate performance relative to AD using the full band set. Experimental results demonstrate that BSS generally performs better background suppression while maintaining target detection capability compared to target detection using full band information.

Hyperspectral image classification faces various levels of difficulty due to the use of different types of hyperspectral image data. Recently, spectral-spatial approaches have been developed by jointly taking care of spectral and spatial information. This paper presents a completely different approach from a subpixel target detection view point. It implements four stage processes, a preprocessing stage, which uses band selection (BS) and nonlinear band expansion, referred to as BS-then-nonlinear expansion (BSNE), a detection stage, which implements constrained energy minimization (CEM) to produce subpixel target maps, and an iterative stage, which develops an iterative CEM (ICEM) by applying Gaussian filters to capture spatial information, and then feeding the Gaussian-filtered CEM-detection maps back to BSNE band images to reprocess CEM in an iterative manner. Finally, in the last stage Otsu's method is applied to converting ICEM-detected real-valued maps to discrete values for classification. The entire process is called BSNE-ICEM. Experimental results demonstrate BSNE-ICEM, which has advantages over support vector machine-based approaches in many aspects, such as easy implementation, fewer parameters to be used, and better false classification and precision rates.

Nowadays, the accidents of oil spill become more and more frequent, causing pollution to the natural resources, marine environment and lives in the sea. As a result, the detection of oil spill draws more and more attentions. One of the most popular region-based active contour models proposed by Chan and Vese, is widely used to image segmentation. But it can't segment hyperspectral oil spill image well, which has blurry boundaries, low distinction, and noise and so on. In order to segment oil spill region from the hyperspectral oil spill image accurately, we improved the region-based active contour model in this paper. For the energy functional, we firstly bring the thought of Fisher criterion into the fitting term to get a better classification result faster. Secondly, a new stop function based on gradient of spectral angle measurement is added into the length term, so as to take advantage of the edge information fully even it is blurry. At last, the model is extended to be able to segment desired material from the complex image with several classes in it. We take some experiments on synthetic and real hyperspectral images to verify the effectiveness of our model, and apply it to the airborne hyperspectral oil spill image. Results of the proposed model on synthetic and testing hyperspectral images show that it outperforms the CV model greatly, and does better than several other segmentation and classification algorithms. Results on hyperspectral oil spill images show that it improves the ability of distinguishing oil spills from sea water, even there are boats and flats in the image.

Recursive band processing of orthogonal subspace projection (RBP-OSP) is developed according to the band sequential (BSQ) format acquired by a hyperspectral imaging sensor. It can be implemented band by band recursively without waiting for data being completely collected. This is particularly important for satellite communication when data download is limited by bandwidth and transmission. Unlike band selection which requires prior knowledge of how many bands are needed to be selected, RBP-OSP has capability which allows different process units to process data whenever bands are available. In addition, it also enables users to identify significant bands during data processing. Finally and most importantly, RBP can provide progressive profiles on OSP performance, which is the best advantage that RBP-OSP can offer and cannot be accomplished by any one-shot operator.

Orthogonal subspace projection (OSP) has found many applications in hyperspectral data exploitation. Its effectiveness and usefulness result from implementation of two stage processes, i.e., annihilation of undesired signal sources by an OSP via inverting a matrix in the first stage followed by a matched filter to extract the desired signal source in the second stage. This paper presents a theory of recursive OSP (ROSP) for hyperspectral imaging, which performs OSP recursively without inverting undesired signature matrices. This ROSP opens up many new dimensions in extending OSP. First of all, ROSP allows OSP to implement varying signatures via a recursive equation without reinverting undesired signature matrices. Second, ROSP can be further used to derive an unsupervised ROSP (UROSP) OSP, which allows OSP to find a growing number of unknown signal sources recursively while simultaneously determining a desired number of signal sources. As a result, the commonly used automatic target generation process (ATGP) can be extended to a recursive ATGP, which can be considered as a special case of UROSP. Third, for practical applications, UROSP can be also extended in two differ ent fashions to causal process and progressive process, which give rise to causal UROSP and progressive UROSP, respectively, both of which can be easily realized in hardware implementation. Finally, UROSP provides a feasible stopping rule via a recently developed UROSP-specified virtual dimensionality.

This paper revisits a well-known fully constrained least squares (FCLS) method developed by Heinz and Chang and develops an approach to finding analytical solutions to FCLS, called analytical FCLS (AFCLS) which can be solved in closed forms instead of FCLS being solved by numerically algorithms. As a result, the AFCLS-unmixed results using analytical solutions are more accurate than FCLS-unmixed results resulting from numerical solutions.

Orthogonal subspace projection (OSP) requires inverting a matrix to eliminate effect of unwanted signal sources on unmixing of desired signal sources. When the number of such wanted signals sources is large, which is indeed the case for hyperspectra data, OSP will become slow due to its matrix inversion. This paper develops a simple alternative approach to OSP without computing matrix inversion, called Gram Schmidt orthogonal vector projection (GSOVP) which is also based on orthogonal projection. Instead of annihilating all unwanted signal sources and then extracting the desired signal as OSP does, GSOVP accomplishes these two tasks by simple inner products. As a result, computational complexity is significantly reduced and hardware design is further simplified.