A novel method for anomaly detection in hyperspectral images (HSIs) is proposed based on low-rank and sparse representation. The the separation of background anomalies observed data. Since each pixel can be approximately represented by a dictionary representation coefficients all pixels form matrix, used to model part. To better characterize pixel's local representation, sparsity-inducing regularization term added coefficients. Moreover, construction strategy adopted make more stable...

In hyperspectral image (HSI) classification, each pixel sample is assigned to a land-cover category. the recent past, convolutional neural network (CNN)-based HSI classification methods have greatly improved performance due their superior ability represent features. However, these limited obtain deep semantic features, and as layer's number increases, computational costs rise significantly. The transformer framework can high-level features well. this article, spectral–spatial feature...

This paper presents a new approach for hyperspectral image classification exploiting spectral-spatial information. Under the maximum posteriori framework, we propose supervised model which includes spectral data fidelity term and spatially adaptive Markov random field (MRF) prior in hidden field. The adopted this is learned from sparse multinomial logistic regression (SMLR) classifier, while MRF modeled by total variation (SpATV) regularization to enforce smooth classifier. To further...

This paper presents a hypserspectral image (HSI) super-resolution method, which fuses low-resolution HSI (LR-HSI) with high-resolution multispectral (HR-MSI) to get (HR-HSI). The proposed method first extracts the nonlocal similar patches form patch tensor (NPT). A novel tensor-tensor product (t - product)-based sparse representation is model extracted NPTs. Through representation, both spectral and spatial similarities between are well preserved. Then, relationship HR-HSI LR-HSI built using...

In recent years, deep learning-based methods have been extensively utilized in remote sensing image scene classification and achieved remarkable performance. The wide geographical coverage resolution differences of images result significant within-class diversity between-class similarity, hindering the further improvement accuracy. Attention-based automatically estimate importance local regions by learning weight assignments, which effectively enhance feature extraction capability network....

This article focuses on hyperspectral image (HSI) super-resolution that aims to fuse a low-spatial-resolution HSI and high-spatial-resolution multispectral form (HR-HSI). Existing deep learning-based approaches are mostly supervised rely large number of labeled training samples, which is unrealistic. The commonly used model-based unsupervised flexible but handcrafted priors. Inspired by the specific properties model, we make first attempt design model-inspired network for in an manner....

The joint use of multisource remote-sensing (RS) data for Earth observation missions has drawn much attention. Although the fusion several sources can improve accuracy land-cover identification, many technical obstacles, such as disparate structures, irrelevant physical characteristics, and a lack training data, exist. In this article, novel dual-branch method, consisting hierarchical convolutional neural network (CNN) transformer network, is proposed fusing heterogeneous information...

The effective combination of hyperspectral image (HSI) and light detection ranging (LiDAR) data can be utilized for land cover classification. Recently, deep learning-based classification methods, especially those utilizing Transformer networks, have achieved remarkable success. However, learning methods multi-source still encounter various technical challenges, such as the comprehensive utilization multi-scale information, lightweight network design, efficient fusion strategies...

In recent years, convolutional neural networks (CNNs) have achieved remarkable success in hyperspectral image (HSI) classification tasks, primarily due to their outstanding spatial feature extraction capabilities. However, CNNs struggle capture the diagnostic spectral information inherent HSI. contrast, vision transformers exhibit formidable prowess handling sequence and excelling at capturing long-range correlations between pixels bands. Nevertheless, loss during propagation, some existing...

Kernel sparse representation classification (KSRC), a nonlinear extension of classification, shows its good performance for hyperspectral image classification. However, KSRC only considers the spectra unordered pixels, without incorporating information on spatially adjacent data. This paper proposes neighboring filtering kernel to spatial-spectral enhanced images. The novelty this work consists in: 1) presenting framework KSRC; and 2) measuring spatial similarity by means neighborhood in...

Anomaly detection plays an important role in remotely sensed hyperspectral image (HSI) processing. Recently, compressive sensing technology has been widely used imaging. However, the reconstruction from HSI and are commonly completed independently, which will reduce processing's efficiency accuracy. In this paper, we propose a framework for with anomaly reconstruct detect anomalies simultaneously. proposed method, is composed of background parts tensor robust principal component analysis...

Hyperspectral (HS) super-resolution, which aims at enhancing the spatial resolution of hyperspectral images (HSIs), has recently attracted considerable attention. A common way HS super-resolution is to fuse HSI with a higher spatial-resolution multispectral image (MSI). Various approaches have been proposed solve this problem by establishing degradation model low HSIs and MSIs based on matrix factorization methods, e.g., unmixing sparse representation. However, category cannot well construct...

Cloud computing offers the possibility to store and process massive amounts of remotely sensed hyperspectral data in a distributed way. Dimensionality reduction is an important task imaging, as often contains redundancy that can be removed prior analysis repositories. In this regard, development dimensionality techniques cloud environments provide both efficient storage preprocessing data. paper, we develop parallel implementation widely used technique for reduction: principal component...

Hyperspectral image (HSI) super-resolution is a hot topic in remote sensing and computer vision. Recently, tensor analysis has been proven to be an efficient technology for HSI processing. However, the existing tensor-based methods of are not able capture high-order correlations HSI. In this article, we propose learn coupled ring (TR) representation super-resolution. The proposed method first tensorizes estimated into which multiscale spatial structures original spectral structure...

This paper presents a new approach for accurate spatial-spectral classification of hyperspectral images, which consists three main steps. First, pixelwise classifier, i.e., the probabilistic-kernel collaborative representation (PKCRC), is proposed to obtain set probability maps using spectral information contained in original data. achieved by means kernel extension based on (CR) classification. Then, an adaptive weighted graph (AWG)-based postprocessing model utilized include spatial...

Kernel methods, e.g., composite kernels (CKs) and spatial-spectral (SSKs), have been demonstrated to be an effective way exploit the information nonlinearly for improving classification performance of hyperspectral image (HSI). However, these methods are always conducted with square-shaped window or superpixel techniques. Both techniques likely misclassify pixels that lie at boundaries class, thus a small target is smoothed away. To alleviate problems, in this paper, we propose novel...

The large amount of data produced by satellites and airborne remote sensing instruments has posed important challenges to efficient scalable processing remotely sensed in the context various applications. In this paper, we propose a new big framework for massive amounts images on cloud computing platforms. addition taking advantage parallel abilities cope with large-scale data, incorporates task scheduling strategy further exploit parallelism during distributed stage. Using computation-...

The large data volume and high algorithm complexity of hyperspectral image (HSI) problems have posed big challenges for efficient classification massive HSI repositories. Recently, cloud computing architectures become more relevant to address the computational introduced in field. This article proposes an acceleration method that relies on scheduling metaheuristics automatically optimally distribute workload applications across multiple resources a platform. By analyzing procedure...

Hyperspectral image super-resolution addresses the problem of fusing a low-resolution hyperspectral (LR-HSI) and high-resolution multispectral (HR-MSI) to produce (HR-HSI). In this paper, we propose novel fusion approach for by exploiting specific properties matrix decomposition, which consists four main steps. First, an endmember extraction algorithm is used extract initial spectral from LR-HSI. Then, with matrix, estimate spatial i.e., spatial-contextual information, degraded observations...

This article gives a survey of state-of-the-art methods for processing remotely sensed big data and thoroughly investigates existing parallel implementations on diverse popular high-performance computing platforms. The pros/cons these approaches are discussed in terms capability, scalability, reliability, ease use. Among distributed platforms, cloud is currently the most promising solution to efficient scalable due its advanced capabilities service-oriented computing. We further provide an...

In the convolutional neural network, precise segmentation of small-scale objects and object boundaries in remote sensing images is a great challenge. As model gets deeper, low-level features with geometric information high-level semantic cannot be obtained simultaneously. To alleviate this problem, successive pooling attention network (SPANet) was proposed. SPANet mainly consists ResNet50 as backbone, module (SPAM) feature fusion (FFM). Specifically, uses two parallel branches to extract by...

As the backbone of Industry 4.0, industrial cyber-physical systems (ICPSs) that are geographically dispersed, federated, cooperative, and security-critical become center interest from both industry academia. In ICPS, there huge amounts devices, such as sensors actuators, which embedded networked together to improve performance real-time monitoring control. Reliability temperature two important concerns these devices in ICPS due their stringent requirement reliable execution long lifespan....

Despite the remarkable progress made by salient object detection of natural sensing images (NSI-SOD), complex background and scale diversity issues remote (RSIs) still pose a substantial obstacle. In this study, we build an end-to-end channel-enhanced remodeling-based network (CRNet) for optical RSIs (ORSIs) to highlight objects through feature augmentation. First, backbone convolutional block is used suggest fundamental characteristics. Then, use channel enhance module (CEM) shallow...

Hyperspectral image (HSI) classification is currently a hot topic in the field of remote sensing. The goal to utilize spectral and spatial information from HSI accurately identify land covers. Convolution neural network (CNN) powerful approach for classification. However, CNN has limited ability capture non-local represent complex features. Recently, vision transformers (ViTs) have gained attention due their process information. Yet, under scenario with ultra-small sample rates,...

Significant progress has been achieved in remote sensing image scene classification (RSISC) with the development of convolutional neural networks (CNNs) and vision transformers (ViT). However, high intra-class diversity inter-class similarity are still enormous challenges for RSISC. Metric learning can effectively improve discriminative ability deep representations by constraining distance between features. Previous metric methods only optimize feature space representation through function,...