Search

Convolutional neural networks (CNNs) have a strong capacity to extract deep-level features from data. However, the standard convolution (SC) only considers the intensity information and ignores the spatial gradient information. Since spatial difference features are more robust to illumination invariance, this letter proposes a multiscale central difference convolutional (MSCDC) network for hyperspectral anomaly detection. Specifically, we use central difference convolution (CDC) to combine intensity and gradient information. This solution improves the representation ability of hyperspectral images (HSIs) and enhances the difference between the background and the anomalies. Furthermore, to fully utilize local spatial information and adapt to targets with different sizes, CDC kernels of three different sizes are used to capture high-, mid-, and low-level features, respectively. Finally, an SC is used to fuse multiscale features and obtain more reliable spatial information. Compared with five popular hyperspectral anomaly detection methods on four real-world HSI datasets, the proposed MSCDC exhibits excellent performance. [ABSTRACT FROM AUTHOR]

Deep learning techniques have shown increasing potential in change detection (CD) in hyperspectral images (HSIs). However, most deep learning-based existing methods for HSI CD follow a patch-based local learning framework and concentrate on binary CD. In this letter, we propose an end-to-end semantic feature fused global learning (SFFGL) framework for HSI multiclass CD (MCD). In SFFGL, a global spatialwise fully convolutional network (FCN), which introduces a spatial attention mechanism (PAM) between encoder and decoder, is designed to effectively exploit the global spatial information from the whole HSIs and achieve patch-free inference. PAM can adaptively extract global spatialwise feature representation. In the model training stage, a global hierarchical (GH) sampling strategy is introduced to obtain diverse gradients during backpropagation for more robust performance. The semantic–spatial feature fusion ($\text{S}^{2}\text{F}^{2}$) unit is designed to effectively fuse the enhanced spatial context information in the encoder and the semantic information in the decoder. More importantly, a semantic feature enhancement module (SEM) is proposed to weaken the influence of the unchanged regional background on the change regional foreground, thus further improving the accuracy. Experimental results on two benchmark HSI datasets demonstrate the effectiveness of the proposed SFFGL. [ABSTRACT FROM AUTHOR]

Hyperspectral images (HSIs) always contain abundant spectral and spatial information. Most of the existing deep learning-based hyperspectral anomaly detection methods consider spectral differences between the background and anomalies, and the local spatial information is usually ignored. To make complete use of the spatial–spectral information, this article proposed a local spatial–spectral information-integrated semisupervised two-stream network (LS3T-Net) for hyperspectral anomaly detection. The two-stream network comprises an adaptive convolution and fully connected network and a variational autoencoder (VAE). The adaptive convolution and fully connected network is used to extract the local spatial features of patches, while VAE is trained to learn spectral information close to the background pixels. Furthermore, the detection maps from the two-stream network are incorporated through a process combining the benefits of spatial learning and spectral learning. This enhances the ability to separate the background and anomalies and suppress the false alarm. The experimental results for six real HSI datasets reveal that LS3T-Net can produce more accurate detection results than seven popular benchmark methods. [ABSTRACT FROM AUTHOR]

Deep learning-based methods (e.g., convolutional neural network (CNN)-based methods) have shown increasing potential in hyperspectral image (HSI) change detection (CD). However, the recent advances in CNN-based methods in HSI CD tasks are mostly devoted to designing more complex architectures or adding additional hand-designed blocks. This increases the number of parameters making model training difficult. In this article, we propose an end-to-end residual self-calibrated network (RSCNet) to increase the accuracy of HSI CD. To fully exploit the spatial information, the proposed RSCNet method adaptively builds interspatial and interspectral dependencies around each spatial location with fewer extra parameters and reduced complexity. Moreover, the introduced self-calibrated convolution (SCConv) helps to generate more discriminative representations by heterogeneously exploiting convolutional filters nested in the convolutional layer. The designed RSC module can explicitly incorporate richer information by introducing response calibration operation. The experiments on four bitemporal HSI datasets demonstrated that the proposed RSCNet method is more accurate than ten widely used benchmark methods. [ABSTRACT FROM AUTHOR]

Due to the limitations of hardware technology and budget constraints, there always exists a tradeoff between spatial and spectral resolutions in a hyperspectral image (HSI). Because of the limited spatial resolution, mixed pixels are a common issue in HSIs, and consequently, some targets appear as subpixels. The effectiveness of hyperspectral target detection is affected greatly by the subpixel targets, especially when the size of the targets is small. In this article, we proposed a double dictionary-based nonlinear representation model for hyperspectral subpixel target detection (DDNRTD). DDNRTD represents HSIs with a nonlinear model based on background and target dictionaries, which fully considers the spatial property of background and targets and can separate background and targets reliably, especially for small-sized subpixel targets. In addition, we designed an over-completed background dictionary construction strategy to represent the background part more effectively, which integrates spectral angle distance (SAD) with sparse representation. Experiments on two simulated and five real HSI datasets showed that the proposed DDNRTD method produced more accurate detection results than six state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Deep learning methods, especially convolutional neural network (CNN)-based methods, have shown promising performance for hyperspectral image (HSI) change detection (CD). It is acknowledged widely that different spectral channels and spatial locations in input image patches may contribute differently to CD. However, they are treated equally in existing CNN-based approaches. To increase the accuracy of HSI CD, we propose an end-to-end Siamese CNN (SiamNet) with a spectral–spatial-wise attention (SSA-SiamNet) mechanism. The proposed SSA-SiamNet method can emphasize informative channels and locations and suppress less informative ones to refine the spectral–spatial features adaptively. Moreover, in the network training phase, the weighted contrastive loss function is used for more reliable separation of changed and unchanged pixels and to accelerate the convergence of the network. SSA-SiamNet was validated using four groups of bitemporal HSIs. The accuracy of CD using the SSA-SiamNet was found to be consistently greater than for ten benchmark methods. [ABSTRACT FROM AUTHOR]

Studies on the classification of hyperspectral images (HSIs) based on deep learning are in full swing, especially the spectral–spatial dependent global learning (SSDGL) framework, which is both efficient and robust. However, the global convolutional long short-term memory (GCL) module under this framework fails to take full consideration of the spectral characteristics contained in HSIs, and the hierarchically balanced (H-B) sampling strategy introduced in this framework prevents the training process from converging smoothly. In this article, we develop a novel regularized spectral–spatial global learning (RSSGL) framework. Compared with SSDGL, the proposed framework mainly makes three improvements. Above all, aiming at the problem that the GCL module used in SSDGL cannot fully tap the local spectral dependence, we apply 3-D convolution to the gated units of long short-term memory (LSTM) as an alternative to the GCL module for adjacent and nonadjacent spectral dependencies learning. Furthermore, to extract the most discriminative features, an improved statistical loss regularization term is developed, in which we introduce a simple but effective diversity-promoting condition to make it more reasonable and suitable for deep metric learning in HSI classification. Finally, to effectively address the performance oscillation caused by the H-B sampling strategy, the proposed framework adopts an early stopping strategy to save and restore the optimal model parameters, making it more flexible and stable. Experiments conducted on three representative datasets show that the proposed RSSGL has superior classification performance compared with the existing relatively excellent research methods. The source code is released at https://github.com/swiftest/RSSGL. [ABSTRACT FROM AUTHOR]

In recent years, convolutional neural networks (CNNs) have achieved excellent performance in hyperspectral image classification and have been widely used. However, the convolution kernel used in traditional CNN has the limitation of single scale, which is not conducive to the improvement of hyperspectral classification performance. In addition, training a classification network of high-dimensional data based on limited labeled samples is still one of the challenges of hyperspectral image classification. To solve the above problems, a hyperspectral image classification method based on expansion convolution network (ECNet) is proposed. The expansion convolution injects holes into the standard convolution kernel to expand the receptive field (RF), so as to extract more context features. Because the shallow features of hyperspectral images contain more location and detail information, while the deep features contain stronger semantic information, in order to further enhance the correlation between deep and shallow information, inspired by ResNet, a similar feedback block (SFB) is introduced on the basis of ECNet, and the deep features and shallow features are fused through this feedback mechanism. Thus, an improved version of ECNet method is obtained, which is called feedback ECNet (FECNet). This study was tested on four commonly used hyperspectral datasets [i.e., Indian Pine (IP), Pavia University (UP), Kennedy Space Center (KSC), and Salinas Valley (SV)] and on a higher resolution and complexly distributed land cover dataset (University of Houston (HT). The experimental results show that the proposed method has better classification performance than some state-of-the-art methods, which shows that FECNet has a certain potential in hyperspectral image classification. [ABSTRACT FROM AUTHOR]

In this letter, a novel subpixel sharpening for soft-then-hard subpixel mapping (SPM) is proposed. First, the fractional images for each class are, respectively, derived by spectral unmixing followed by spatial interpolation and by spectral interpolation followed by spectral unmixing. Bilinear and bicubic interpolation is used as the spatial and spectral interpolation methods. The fractional images for each class are then integrated together using the appropriate weighting parameter. Finally, the integrated finer fractional images are used to allocate hard class labels to subpixels. The proposed method is fast and does not need any prior spatial structure information. Experiments on two actual hyperspectral images show that the proposed method produces higher accuracy results than the existing algorithms. Moreover, both the spatial and spectral information is fully utilized to improve the accuracy of the SPM results. [ABSTRACT FROM PUBLISHER]

The subpixel mapping (SPM) technique can handle coarse fractional images derived by unmixing coarse original hyperspectral (HS) image to produce a fine land cover map at the subpixel scale. A popular SPM approach is a two-step model. It first increases the spatial resolution of coarse fractional images by subpixel sharpening to produce fine fractional images and then assigns class labels to each subpixel by the class allocation method. However, there is only a single processing path of the current SPM algorithm, and the information type of the fine fractional images is not rich. To enrich the information type, SPM based on dual processing paths (DPP) is proposed. DPP contains two processing paths, namely spatial–spectral path and multiscale path. First, the coarse original HS image and the high spatial resolution multispectral image are fused by component substitution to produce the fine fractional images with more spatial–spectral information in the spatial–spectral path. At the same time, deep Laplacian pyramid networks are used to obtain the fine fractional images with multiscale information in the multiscale path. The fine fractional images from the two paths are then integrated to generate the improved fraction images with multiscale spatial–spectral information. Finally, the multiscale spatial–spectral information is utilized to allocate class labels by the class allocation method. Experimental results on three real HS remote sensing data show that the proposed DPP outperforms the other SPM methods, demonstrating the effectiveness of the use of DPP in enriching the information type of the fine fractional images. [ABSTRACT FROM AUTHOR]

Multiple subpixel shifted images (MSIs) from the same area can be incorporated to improve the accuracy of soft-then-hard subpixel mapping (STHSPM). In this paper, a novel method that derives higher resolution MSIs with more spatial–spectral information (MSI-SS) is proposed. First, coarse MSIs produce two high-resolution MSIs for each class respectively by two paths at the same time. The spatial path produces the high-resolution MSIs by soft classification followed by interpolation. And, the other high-resolution MSIs are derived from the spectral path by interpolation followed by soft classification. Then the higher resolution MSIs with more spatial–spectral information for each class are derived by integrating the aforementioned two kinds of high-resolution MSIs by the appropriate weight. Finally, the integrated higher resolution MSIs for each class are used to allocate hard class labels to subpixels. The proposed method is fast and takes more spatial–spectral information of the original MSIs into account. Experiments on three real hyperspectral remote sensing images show that the proposed method produce higher SPM accuracy result. [ABSTRACT FROM PUBLISHER]

Subpixel mapping (SPM) of hyperspectral remote sensing imagery is a promising technique for deriving fine mapping result by classification at fine spatial resolution. There is a type of algorithm for SPM, namely, the soft-then-hard SPM (STHSPM) algorithm that first estimates soft attribute values for land cover classes at subpixel level and then allocates classes for subpixel according to the soft attribute values. However, the fraction images derived from spectral unmixing are of less prior information of original hyperspectral remote sensing imagery and there are lots of errors in SPM result due to the limitation of spectral unmixing technology currently available. In this paper, a framework based on subpixel resolution thematic map, namely, super-resolution then classification (STC) is proposed to improve mapping result. In the proposed framework, a maximum a posteriori (MAP) model associated with the endmembers of interest (EOI), namely, T-MAP-SR is applied to the original coarse imagery to derive a high-resolution imagery with generous prior information. Then fine mapping result can be derived from the high-spatial resolution imagery by the available classification methods. Experiments show that the proposed framework can produce higher mapping accuracy result and protect the classes of interest (COI). [ABSTRACT FROM PUBLISHER]

Due to the limited spatial resolution of multispectral/hyperspectral data, mixed pixels widely exist and various spectral unmixing techniques have been developed for information extraction at the subpixel level in recent years. One of the challenging problems in spectral mixture analysis is how to model the data of a primary class. Given that the within-class spectral variability (WSV) is inevitable, it is more realistic to associate a group of representative spectra with a pure class. The unmixing method using the extended support vector machines (eSVMs) has handled this problem effectively. However, it has simplified WSV in the mixed cases. In this paper, a further development of eSVMs is presented to address two problems in multiple-endmember spectral mixture analysis: 1) one mixed pixel may be unmixed into different fractions (model overlap); and 2) one fraction may correspond to a group of mixed pixels (fraction overlap). Then, spectral unmixing resolution (SUR) is introduced to characterize how finely the mixture in a mixed pixel can be quantified. The quantitative relationship between SUR and WSV of endmembers is derived via a geometry analysis in support vector machine feature space. Thus, the possible SUR can be estimated when multiple endmembers for each class are given. Moreover, if the requirement of SUR is fixed, the acceptance level of WSV is then limited, which can be used as a guide to remove outliers and purify endmembers for each primary class. Experiments are presented to illustrate model and fraction overlap problems and the application of SUR in uncertainty analysis of spectral unmixing. [ABSTRACT FROM PUBLISHER]

The kernel function plays an important role in machine learning methods such as the support vector machine. In this letter, a new kernel framework is developed for hyperspectral image classification. In contrast to existing composite kernels constructed via a linearly weighted combination, the multiple-mapping kernel proposed in this letter is obtained through repeated nonlinear mappings. Experiments indicate that the proposed multiple-mapping kernel framework (MMKF) is effective for hyperspectral image classification. Compared to the single kernel methods, the MMKF tends to be more advantageous in terms of classification accuracy, particularly for the situation with a small-size training set. [ABSTRACT FROM PUBLISHER]

Spectral unmixing is one of the important techniques for hyperspectral data processing. The analysis of spectral mixing is often based on a linear, fully constrained (FC) (i.e., nonnegative and sum-to-one mixture proportions), and least squares criterion. However, the traditional iterative processing of FC least squares (FCLS) linear spectral mixture analysis (LSMA) (FCLS-LSMA) is of heavy computational burden. Recently developed geometric LSMA methods decreased the complexity to some degree, but how to further reduce the computational burden and completely meet the FCLS criterion of minimizing the unmixing residual needs to be explored. In this paper, a simple distance measure is proposed, and then, a new geometric FCLS-LSMA method is constructed based on the distance measure. The method is in line with the FCLS criterion, free of iteration and dimension reduction, and with very low complexity. Experimental results show that the proposed method can obtain the same optimal FCLS solution as the traditional iteration-based FCLS-LSMA, and it is much faster than the existing spectral unmixing methods, particularly the traditional iteration-based method. [ABSTRACT FROM AUTHOR]

Endmember extraction (EE) is a prerequisite task for spectral analysis of hyperspectral imagery. In all kinds of EE algorithms, maximum simplex volume-based ones, such as simplex growing algorithm (SGA) and N-FINDR algorithm, have been widely used for their fully automated and efficient performance. However, implementation of the algorithms needs dimension reduction of original data, and the algorithms include innumerable volume calculation. This leads to a low speed of the algorithms and thus becomes a limitation to their applications. In this paper, a simple distance measure is presented, and then, fast SGA and fast N-FINDR algorithm are constructed based on a proposed distance measure, which is free of dimension reduction and makes use of distance measure instead of volume evaluation to speed up the algorithm. The complexity of the proposed methods is compared with the original algorithms by theoretical analysis. Experiments show that the implementation of the two improved EE algorithms is much faster than that of the two original maximum simplex volume-based EE algorithms. [ABSTRACT FROM PUBLISHER]