Hyperspectral image classification (HSIC) plays a key role in remote sensing, but the interpretation of scene features with limited samples remains challenging. In this letter, we propose a multiscale spectral morphological fusion network (MSMFNet) for unsupervised domain adaptation (UDA) in HSIC. This method consists of several key steps. First, the network uses principal component analysis (PCA) to extract spectral features, which are then combined with extended morphological profiles (EMPs) to enhance spatial structure representation. Then, a multiscale heterogeneous feature aggregation (MHFA) module is introduced to capture heterogeneous features across different scales and directions. Next, the multiscale global attention (MGA) module generates multilevel responses, exploring correlations between local and global information, thereby improving the representation of fine-grained features and boosting cross-domain feature fusion and classification performance. Finally, contrastive learning (CL) is applied to efficiently extract domain-invariant features. The experimental results demonstrate that MSMFNet achieves superior performance in cross-domain adaptation and fine-grained feature discrimination, achieving accuracies of 77.48% on the Houston dataset and 93.66% on the Pavia dataset, with Kappa coefficients that exceed the state of the art by 2.46 and 2.48, respectively.

Deep neural networks play a significant role in hyperspectral image (HSI) processing, yet they can be easily fooled when trained with adversarial samples (generated by adding tiny perturbations to clean samples). These perturbations are invisible to the human eye, but can easily lead to misclassification by the deep learning model. Recent research on defense against adversarial samples in HSI classification has improved the robustness of deep networks by exploiting global contextual information. However, available methods do not distinguish between different classes of contextual information, which makes the global context unreliable and increases the success rate of attacks. To solve this problem, we propose a robust context-aware network able to defend against adversarial samples in HSI classification. The proposed model generates a global contextual representation by aggregating the features learned via dilated convolution, and then explicitly models intraclass and interclass contextual information by constructing a class context-aware learning module (including affinity loss) to further refine the global context. The module helps pixels obtain more reliable long-range dependencies and improves the overall robustness of the model against adversarial attacks. Experiments on several benchmark HSI datasets demonstrate that the proposed method is more robust and exhibits better generalization than other advanced techniques.

To address the conflict caused by the fixed sampled period in the multi-sampled multiband-structured subband adaptive filter (MS-MSAF), the adaptive convex combination of two MS-MSAFs is proposed in this paper, in which the individual filters independently run with different sampled periods. Moreover, the mean behavior of the convex-combined two MS-MSAF algorithm is also studied. In addition, the convex-combined two MS-MSAF algorithm with periodic feedback is also developed to further enhance the convergence performance. Finally, the validity of the proposed adaptive filtering algorithms together with their theoretical analyses is supported by computer simulations.

The affine combination of two complex-valued least-mean-squares filters (aff-CLMS) addresses the trade-off between fast convergence rate and small steady-state misadjustment error. However, a rigorous analysis of the aff-CLMS algorithm for second-order non-circular inputs is still under investigation. To this end, the focus in this letter is on the full mean-square analysis of the aff-CLMS algorithm, in which the transient analyses of the mixing parameter, as well as the standard and complementary weight-error covariance matrices, are completed. In addition, we derive the closed-form solutions of the steady-state weight-error power and its complementary version of the aff-CLMS. Finally, the effectiveness of the theoretical analysis is supported by computer simulations.

This paper introduces two combined-sample multiband-structured subband adaptive filters (MSAFs). In the design, an adaptive convex combination scheme of two self-reliant multi-sampled MSAF (MS-MSAF) with different sampled periods is firstly developed, which leads to the so-called CTMS-MSAF algorithm. Secondly, based on an adaptive filter, the combined-sample MS-MSAF (CMS-MSAF) algorithm is proposed via designing a time-varying sampled period, which possesses lower computational complexity than the former. Then, the convergence behaviors of the CTMS-MSAF and CMS-MSAF algorithms are investigated using standard mean-square deviation analysis. Finally, the simulation study in the system identification and acoustic echo cancellation applications shows that at the same steady-state error, the CMS-MSAF method provides a faster convergence rate than the improved convex combination of two MSAFs, combined-step-size MSAF and CTMS-MSAF algorithms.

Convolutional neural networks (CNNs) have been widely used in hyperspectral image (HSI) classification tasks because of their excellent local spatial feature extraction capabilities. However, because it is difficult to establish dependencies between long sequences of data for CNNs, there are limitations in the process of processing hyperspectral spectral sequence features. To overcome these limitations, inspired by the Transformer model, a spatial–spectral transformer with cross-attention (CASST) method is proposed. Overall, the method consists of a dual-branch structures, i.e., spatial and spectral sequence branches. The former is used to capture fine-grained spatial information of HSI, and the latter is adopted to extract the spectral features and establish interdependencies between spectral sequences. Specifically, to enhance the consistency among features and relieve computational burden, we design a spatial–spectral cross-attention module with weighted sharing to extract the interactive spatial–spectral fusion feature intra Transformer block, while also developing a spatial–spectral weighted sharing mechanism to capture the robust semantic feature inter Transformer block. Performance evaluation experiments are conducted on three hyperspectral classification datasets, demonstrating that the CASST method achieves better accuracy than the state-of-the-art Transformer classification models and mainstream classification networks.

Hyperspectral images (HSIs) contain abundant information in the spatial and spectral domains, allowing for a precise characterization of categories of materials. Convolutional neural networks (CNNs) have achieved great success in HSI classification, owing to their excellent ability in local contextual modeling. However, CNNs suffer from fixed filter weights and deep convolutional layers, which lead to a limited receptive field and high computational burden. The recent Vision Transformer (ViT) models long-range dependencies with a self-attention mechanism and has been an alternative backbone to the CNNs traditionally used in HSI classification. However, such transformer-based architectures designate all input pixels of the receptive field as feature tokens in terms of feature embedding and self-attention, which inevitably limits the ability for learning multi-scale features and increases the computational cost. To overcome this issue, we propose a local semantic feature aggregation-based transformer (LSFAT) architecture which allows transformers to represent long-range dependencies of multi-scale features more efficiently. We introduce the concept of the homogeneous region into the transformer by considering a pixel aggregation strategy and further propose neighborhood aggregation-based embedding (NAE) and attention (NAA) modules, which are able to adaptively form multi-scale features and capture locally spatial semantics among them in a hierarchical transformer architecture. A reusable classification token is included together with the feature tokens in the attention calculation. In the last stage, a fully connected layer is employed to perform classification on the reusable token after transformer encoding. We verify the effectiveness of the NAE and NAA modules compared with the traditional ViT through extensive experiments. Our results demonstrate the excellent classification performance of the proposed method in comparison with other state-of-the-art approaches on several public HSIs.

This letter introduces a novel spatial-spectral classification method for hyperspectral images (HSIs) based on a structural-kernel collaborative representation (SKCR), which considers one weak assumption of spatial neighborhood that of the pixels in a superpixel belong to the same class when exploiting contextual information in HSI. The proposed method consists of the following steps. First, a superpixel segmentation strategy is used to construct self-adaptive regions for the HSI. Then, the structural information within each superpixel block is extracted based on the density peak and K nearest neighbors. Next, dual kernels are separately utilized for the exploitation of the spectral and the spatial information. Finally, the dual kernels are combined and incorporated into a support-vector-machine classifier. Since the weak assumption of spatial neighborhood is well considered in the collaborative representation, the proposed method showed excellent classification performance for two widely used real hyperspectral data sets even when the number of training samples was relatively small.

Spectral–spatial information plays an essential role in hyperspectral image (HSI) classification compared to pure spectral information. However, the neighbor spectral–spatial information of a pixel tends to be mixed into other ground coverings due to various external factors such as the weather and sensor jitter, and mainstream HSI classification methods present low sensitivity for spatial information in this situation. This article proposes a novel feature extraction method via 3-D block characteristics sharing (3-D-BCS) for HSI classification that redefines spatial–spectral information of a local region based on a superpixel perspective to overcome the spectral–spatial weak assumptions in feature extraction that consists of the following steps. First, 3-D blocks are obtained by performing an oversegmentation method on the raw HSI. Then, instead of global operation, a 3-D block-based Gabor filter is applied to the principal components of an HSI to extract the textural features. Next, an average operation is conducted on each shape adaptive region to address the spatial weak assumption and Gaussian weight is introduced into each superpixel block to overcome the spectral weak assumption. Thus, 3-D characteristics sharing blocks can be constructed by reshaping the above three kinds of spectral–spatial feature. Finally, the majority-based support vector machine (SVM) classifier is utilized to determine the final class labels of HSI at the decision fusion level. Experiments performed on several real hyperspectral data sets with limited training samples show that the proposed 3-D-BCS method outperforms the other types of the classification method.