The application of graph convolutional networks (GCNs) to hyperspectral image (HSI) classification is a heavily researched topic. However, GCNs are based on spectral filters, which computationally costly and fail suppress noise effectively. In addition, the current GCN-based methods prone oversmoothing (the representation each node tends be congruent) problems. To circumvent these problems, novel semi-supervised locality-preserving dense neural network (GNN) with autoregressive moving...

With limited number of labeled samples, hyperspectral image (HSI) classification is a difficult Problem in current research. The graph neural network (GNN) has emerged as an approach to semi-supervised classification, and the application GNN images attracted much attention. However, existing GNN-based methods single or filter mainly used extract HSI features, which does not take full advantage various networks (graph filters). Moreover, traditional GNNs have problem oversmoothing. To...

Recently, graph convolutional network (GCN) has achieved promising results in hyperspectral image (HSI) classification. However, GCN is a transductive learning method, which difficult to aggregate the new node. Besides, existing GCN-based methods divide construction and classification into two stages ignoring influence of constructed error on results. Moreover, available fail understand global contextual information graph. In this article, we propose novel multiscale sample with...

Due to prior knowledge deficiency, large spectral variability, and high dimension of hyperspectral image (HSI), HSI clustering is extremally a fundamental but challenging task. Deep methods have achieved remarkable success attracted increasing attention in unsupervised classification (HSIC). However, the poor robustness, adaptability, feature presentation limit their practical applications complex large-scale datasets. Thus, this article introduces novel self-supervised locality preserving...

Hyperspectral image (HSI) clustering is an extremely fundamental but challenging task with no labeled samples. Deep methods have attracted increasing attention and achieved remarkable success in HSI classification. However, most existing are ineffective for large-scale HSI, due to their poor robustness, adaptability, feature presentation. In this paper, address these issues, we introduce unsupervised self-correlated learning smoothy enhanced locality preserving graph convolution embedding...

Nowadays, automatic modulation classification (AMC) has become a key component of next-generation drone communication systems, which are crucial for improving efficiency in non-cooperative environments. The contradiction between the accuracy and current methods hinders practical application AMC systems. In this paper, we propose real-time method based on lightweight mobile radio transformer (MobileRaT). constructed is trained iteratively, accompanied by pruning redundant weights information...

Graph convolutional network (GCN) has shown potential in hyperspectral image (HSI) classification. However, GCN is a transductive learning method, which difficult to aggregate the new node. The available GCN-based methods fail understand global and contextual information of graph. To address this deficiency, novel semisupervised based on graph sample aggregate-attention (SAGE-A) for HSIs' classification proposed. Different from SAGE-A adopts multilevel (graphSAGE) network, as it can flexibly...

ABSTRACTRecently, the research area of hyperspectral (HS) image change detection (CD) is popular with convolutional neural networks (CNNs) based methods. However, conventional CNNs-based CD algorithms commonly achieve by comparing deep features extracted from bi-temporal images at decision level, which often fails to take full advantage network different levels. Moreover, there are inevitably substantial redundant located in non-varying regions images, considerably impedes training...

Self-supervised hyperspectral image (HSI) clustering remains a fundamental yet challenging task due to the absence of labeled data and inherent complexity spatial-spectral interactions. While recent advancements have explored innovative approaches, existing methods face critical limitations in accuracy, feature discriminability, computational efficiency, robustness noise, hindering their practical deployment. In this paper, self-supervised efficient low-pass contrastive graph (SLCGC) is...

Hyperspectral image (HSI) clustering has been a fundamental but challenging task with zero training labels. Currently, some deep graph methods have successfully explored for HSI due to their outstanding performance in effective spatial structural information encoding. Nevertheless, insufficient utilization, poor feature presentation ability, and weak update capability limit performance. Thus, this paper, homophily structure learning an adaptive filter method (AHSGC) is proposed....

Seizure onset detection and epileptic preictal prediction based on electroencephalogram (EEG) signals have been a challenge problem in the research community. In this brief, novel states classification algorithm multichannel EEGs representation using multiple hand-crafted features, feature fusion transfer learning (TL) with pre-trained deep neural networks (DNNs), discriminative extraction state hierarchical network (HNN), is developed. The mean amplitude spectrum (MAS), power spectral...

Deep subspace clustering has achieved remarkable performances in the unsupervised classification of hyperspectral images. However, previous models based on pixel-level self-expressiveness data suffer from exponential growth computational complexity and access memory requirements with increasing number samples, thus leading to poor applicability large This paper presents a Neighborhood Contrastive Subspace Clustering network (NCSC), scalable robust deep approach, for Instead using...

Transformer has been widely used in classification tasks for hyperspectral images (HSI) recent years. Because it can mine spectral sequence information to establish long-range dependence, its performance be comparable with the convolutional neural network (CNN). However, both CNN and focus excessively on spatial or domain features, resulting an insufficient combination of spatial-spectral from HSI modeling. To solve this problem, we propose a new end-to-end graph (GCN) fusion feature...

The joint clustering of multimodal remote sensing (RS) data poses a critical and challenging task in Earth observation. Although recent advances multiview subspace have shown remarkable success, existing methods become computationally prohibitive when dealing with large-scale RS datasets. Moreover, they neglect intrinsic nonlinear spatial interdependencies among heterogeneous lack generalization ability for out-of-sample data, thereby restricting their applicability. This article introduces...

Encouraging progress in few-shot semantic segmentation has been made by leveraging features learned upon base classes with sufficient training data to represent novel examples. However, this feature sharing mechanism inevitably causes aliasing between when they have similar compositions of concepts. In paper, we reformulate as a reconstruction problem, and convert class into series basis vectors which span class-level space for reconstruction. By introducing contrastive loss, maximize the...

Nuclear instance segmentation is a challenging task due to large number of touching and overlapping nuclei in pathological images. Existing methods cannot effectively recognize the accurate boundary owing neglecting relationship between pixels (e.g., direction information). In this paper, we propose novel Centripetal Direction Net-work (CDNet) for nuclear segmentation. Specifically, define centripetal feature as class adjacent directions pointing center rep-resent spatial within nucleus....