Search

Remote sensing technology in the new era gradually need a new scientific paradigm driven by big data and knowledge. However, in the current research, the use of existing knowledge is slightly insufficient. As for the whole research process, there is a lack of guidance based on the field knowledge. In view of the above problems in the field of remote sensing in agriculture, it is of great significance to organize knowledge based on concept graph and develop knowledge-driven big data analysis theory and methods. This paper focuses on the construction and application of agricultural remote sensing concept graph. Firstly, it explores the core contents, research hotspots and development trends of the current agricultural remote sensing field through reference visualization and analysis methods such as keyword frequency statistics, keyword co-occurrence, word detection with stronger citation burst and so on. Then, this paper summarizes the comprehensive ontology of agricultural remote sensing based on the results of keyword clustering. Thus, the schema layer of concept graph is constructed from top to bottom. Under the guidance of the schema layer, the data layer of the concept graph is artificially filled from the bottom up with different triads stored in Neo4j. In terms of application, retrieval, query and recommendation of the proposed concept graph for agricultural remote sensing tasks is a typical application case. Based on the results of the recommendation, it could guide the process of information extraction such as cultivated land parcel delineation and crop type identification at the parcel scale. The study of concept graph helps to summarize and generalize the relevant knowledge in the field of agricultural remote sensing. Relevant research provide effective solutions for automated and intelligent agricultural information extraction and application. [ABSTRACT FROM AUTHOR]

Potash plays an important role in agriculture and food security, so does the estimation of the content of substances in potash. Chemical methods for such estimation tasks are time-consuming, whereas thermal infrared hyperspectral (TIH) images provide us with an opportunity to do estimation efficiently. This paper hence applies the TIH technology to predict the content of various substances in potash. On the one hand, the TIH technology provides rich hyperspectral information for substance discovery and analysis. On the other hand, it introduces challenges of dimensionality and information redundancy, leading to potential decreased predicting accuracy. To solve the problem, as well as easing the burden of detection instruments in terms of band usage, band selection is critical. This paper proposes a dynamic metric-guided particle swarm optimization (DM-GPSO) method for band selection. It adaptively adjusts parameters to prevent local optima, avoiding the need for manual parameter tuning. Variety of band selection algorithms, including the genetic algorithm (GA), particle swarm optimization (PSO), and improved PSO algorithms, are tested to extract and analyse the characteristic bands of potash. Experimental results demonstrate that the TIH technology combined with a proper band selection methodis effective in estimating the content of potash. The new DM-GPSO can select different combinations of bands, which lead to better and more stable prediction results. The analysis on the combinations of selected bands also reveals that picromerite and potassium chloride have similar intervals of characteristic bands, which can be used to guide the production process of potash in practice. [ABSTRACT FROM AUTHOR]

In the context of global carbon emission reduction, solar photovoltaic (PV) technology is experiencing rapid development. Using high-resolution remote sensing images to accurately obtain PV information over a large region, including location and size, has the advantages of high statistical efficiency and timely data update for the PV energy management. Due to the intra-class diversity of PV panels and the intricate variability in their deployment environments, existing semantic segmentation methods often have problems such as under-segmentation and mis-segmentation. To alleviate these problems, this paper proposes an improved DeepLabv3+ semantic segmentation network to more accurately extract PV panels from high-resolution remote sensing images. With the aim of alleviating under-segmentation, a multi-level context aggregation module is developed. This module can enhance the model's ability to learn the characteristics of PV panels and their surrounding environment by aggregating rich contextual information from multi-scale and semantic levels. To alleviate the problem of mis-segmentation, a hybrid attention module is introduced. This module sequentially and adaptively adjusts the weight distribution in both the channel and spatial dimensions, thus enabling the model to focus more on the feature information and spatial positions of PV objects. Experiments conducted on a self-constructed Beijing PV segmentation dataset show that the method in this paper has advantages of completeness and accuracy in extracting PV panels compared to the baseline model and current mainstream semantic segmentation network. In addition, the results of experiments on extracting PV panels in real region show that our model also has good stability and generalization capability. [ABSTRACT FROM AUTHOR]

In recent years, low-rank sparse representation (LRR) schemes have been applied to the field of hyperspectral image (HSI) anomaly detection (AD) and received attention. However, it is often accompanied by a small amount of noise in the AD, which leads to a high false alarm rate of detection results. In this paper, the HSI-AD algorithm is proposed to address this problem. First, the HSI is unmixed using the deep network to obtain the endmembers and abundance, and the abundance matrix is fused with the HSI to obtain the mixed data, which to a certain extent complements the spatial information of the HSI because the abundance matrix can reflect the subpixel-level information of the target image elements. Then, the mixed data are decomposed based on the model proposed in this paper. To overcome the situation that the sparse part of the traditional model is prone to mix with a small amount of noise, the sparse decomposition of the mixed data into anomaly and noise is used to remove the small amount of noise from the sparse part. A joint dictionary consisting of background and anomaly is used in the above model. In addition, to address the problem of information loss caused by the uniform accumulation of kernel norm over all singular values in the process of model decomposition, this paper designs the γ norm decomposition method to replace the general kernel norm decomposition for a more effective extraction of useful information. [ABSTRACT FROM AUTHOR]

Point cloud registration is the process of aligning and merging multiple point clouds into a same coordinate system. It has many applications in computer vision, robotics, 3D reconstruction, augmented reality, and autonomous driving. Despite its wide-ranging utility, traditional algorithms grapple with challenges such as substantial noise, outliers, limited overlap, motion distortions, and handling large scenes. These complexities impede the effective deployment of point cloud registration in practical real-world scenarios. This paper conducts a comprehensive inquiry into the domain of deep learning-based point cloud registration. It synthesizes recent advancements in algorithms about point cloud registration grounded in deep learning. The exposition delineates insights from four distinct perspectives: multi-layer perceptron, deep neural network, graph convolutional network, and attention mechanism. Furthermore, the paper systematically reviews prevalent datasets and evaluation metrics associated with deep learning-based point cloud registration. A discussion on the existing challenges in point cloud registration and a forward-looking exploration of potential research directions are included. [ABSTRACT FROM AUTHOR]

Leaf area index (LAI) assessment methods relying on physical and empirical models are considered to be the most commonly used method at present, but their estimation efficiency and accuracy are deficient. Although the hybrid model of these two methods can address these issues, a poor coupling mechanism can easily occur. Given that, a PROSAIL model coupling particle swarm optimization (PSO) neural network (NN) algorithm (PSO-NN-PROSAIL model) was introduced to invert the winter wheat LAI (WWLAI) at five distinct growth stages. The Xiangfu District in the east of Kaifeng City, Henan Province was served as the study region. Based on the measured WWLAI data at varying stages and GF-1 WFV satellite images, the initial analysis focused on assessing the PROSAIL model's sensitivity in simulating vegetation canopy reflectance. It then calculated six vegetation index models according to the wavelength reflectance of GF-1 WFV and analysed their correlation with LAI to select the input parameters that could be used in the model. Normalized differential vegetation Index (NDVI) and ratio vegetation Index (RVI) as well as vegetation canopy reflectance were employed as input parameters to invert the WWLAI by adopting the PSO-NN-PROSAIL model. The experimental results showed the following: (1) In the vegetation index model, the determination coefficient (R2) of NDVI and RVI was greater than 0.68, implying that NDVI and RVI might serve as input factors for the proposed model in this paper; (2) LAI and chlorophyll a + b content (Cab) were most sensitive to the PROSAIL model in near-infrared and visible light bands; and (3) the PSO-NN-PROSAIL model possessed better LAI inversion accuracy. In summary, the model proposed in this paper provided a technical reference for rapid and accurate remote sensing monitoring of WWLAI. The optimization technology of particle swarm optimization algorithm is integrated into the neural network, and the weight of the neural network is adjusted by its fitness function transformation and inertia weight. By integrating prior knowledge into the inversion of crop leaf area index constructed by machine learning and radiative transfer model, the ill-posed problem of physical model inversion has been improved. [ABSTRACT FROM AUTHOR]

Seagrass beds play a vital ecological role by regulating climate, maintaining biodiversity, and sequestering carbon. Achieving accurate classification of seagrass beds across both time and space is of significant importance for efficient monitoring. This paper addresses the challenges of seagrass bed classification in remote sensing imagery and the difficulty of achieving temporal and spatial transferability of classification model. Leveraging a long-term Landsat satellite image time series spanning from 1976 to 2022 in the Yellow River Delta, we enhance the DeepLab V3+ network and propose a cross-temporal-spatial domain seagrass bed transfer learning classification approach. We utilize selectively chosen images from a few years in partial regions for training, introduce freeze-training and-unfreeze-training during the model training phase, and then transfer the model to achieve cross-temporal and cross-temporal-spatial domain classification on images from other years and regions. Additionally, we discuss the factors influencing the classification results. The research results indicate that: (1)The method we proposed has shown good results, achieving classification accuracy of 83.17% and 80.99% for seagrass beds in both cross-temporal and cross-temporal-spatial domains. (2) The classification results indicate that this method achieves an average classification accuracy higher by 16.18% compared to networks such as UNet, PSPNet, HRNet, UNet++, SegFormer, etc. under cross-temporal conditions and higher by 17.56% under cross-spatiotemporal conditions for seagrass bed classification. (3) Based on the classification results, it is observed that from 1973 to 1986, the overall area of seagrass beds in the Yellow River Delta had an initial growth trend, with the largest seagrass bed area recorded in 1986, reaching 4281.289 hectares. Subsequently, the seagrass beds showed an overall decreasing trend, and by 2019 they had almost completely disappeared. The cross-temporal-spatial domain seagrass bed transfer learning classification method proposed in this paper can provide valuable support for the monitoring, restoration, management, and utilization of seagrass ecosystems. [ABSTRACT FROM AUTHOR]

Classification of hyperspectral image (HSI) is an essential step in various remote sensing tasks. Currently, transformer-based models have made significant contributions to HSI classification due to their remarkable ability to model the relation among long-range sequence data. However, HSI data exhibit strong nonlinear coupling and strong correlation, leading to a decrease in the classification performance of transformer-based models. The SpectralFormer network has achieved excellent performance in HSI classification, but its cross-layer adaptive fusion mechanism cannot effectively highlight critical information, affecting the representation ability of spectral-spatial features. To address the above issues, this paper proposes an HSI classification method based on local feature decoupling and SpectralFormer with a hybrid attention module (HAMSF) network. Specifically, this paper first uses the histogram of oriented gradient algorithm to preprocess HSI coupled data and achieve preliminary nonlinear decoupling of HSI data. Secondly, the hybrid attention module is embedded in a cross-layer adaptive fusion mechanism, generating refined neighbouring information before each information fusion to improve the spectral-spatial feature representation ability. Finally, the processed data are input into the HAMSF network for the final decoupling and classification of HSI. The experiment is carried out on three benchmark hyperspectral datasets, and the experimental results confirm that the proposed method has better classification performance compared to several transformer-based and classical HSI classification methods. [ABSTRACT FROM AUTHOR]

Initialization plays an important role in the accuracy of endmember extraction algorithms (EEAs) in linear hyperspectral unmixing (LHU). Random initialization can lead to varying endmembers generated by EEAs. To address this challenge, an initialization strategy has been introduced, encompassing vertex component analysis (VCA), automatic target generation process (ATGP), among others. These techniques significantly contribute to enhancing the accuracy of EEAs. However, complex initialization is sometimes less preferable, prompting the unexplored question of whether there exists an EEA robust to initialization. This paper focuses on analyzing this issue within the context of minimum simplex volume-based (MV) methods, which have received considerable attention in the past two decades due to their robustness against the absence of pure pixels. MV methods typically formulate LHU as an optimization problem, most of which includes a non-convex volume term. Additionally, many MV methods use VCA as an initialization strategy. Firstly, this paper demonstrates that the variable splitting augmented Lagrangian approach (SISAL), as a representative non-convex MV method, heavily depends on initialization. To our knowledge, the impact of initialization for MV methods has not been thoroughly analyzed before. Furthermore, this paper proposes an initially robust MV method by introducing a new convex MV term. Numerical experiments conducted on simulated and real datasets demonstrate its outstanding performance in accuracy and robustness to initialization. Throughout the experiments the proposed method proves to be the most stable, which is crucial in real scene where the ground truth is unknown beforehand. [ABSTRACT FROM AUTHOR]

Classification with a few samples of training set has been a long-standing issue in the field of polarimetric synthetic aperture radar (PolSAR) image analysis and processing. In fact, one of the most important challenges of the PolSAR image classification task is the number of labelled samples. In essence, training network by utilizing just few numbers of training samples cannot be led to the reliable result since the neural network is so sensitive to the number of training samples. In addition, in the real PolSAR image classification task, the huge number of training samples is not accessible. On the other hand, classification performance using just small number of labelled samples is not an accurate result. So, in this paper, aiming at the small number of training samples of the PolSAR image classification task, a novel ensemble self-supervised feature-learning (ESSFL) model is designed. The designed ESSFL can automatically extract PolSAR features conducive to PolSAR image classification with a small number of training samples. In addition, it can significantly decrease the dependence of neural network algorithms on large labelled samples of training set. First, to utilize the spatial–polarimetric features of PolSAR data perfectly, the EfficientNet-B0 is presented and utilized as the main section of the deep learning (DL) model to extract DL features of PolSAR data. Then, using an optimization function that constrains the cross-correlation matrix of various distortions of each sample to the identity matrix, the designed deep learning model can obtain the effective features of homogeneous samples gathering together and heterogeneous samples separating from each other in a self-supervised manner. Moreover, two ensemble learning (EL) models, feature-level and view-level ensemble, are proposed to increase the feature extraction capability and classification result by jointly using spatial features at different scales and polarimetric information at different bands. Finally, the stack of the obtained features and the main polarimetric information of PolSAR data can be input into classifier for the classification of PolSAR data. In this paper we use two different classifiers, the first one is the Random Forest (RF) classifier and the second one is the Support Vector Machine (SVM) classifier. The advantages of SVM include that it can be used to avoid the difficulties of using linear functions in the high-dimensional feature space, and the optimization problem is transformed into dual convex quadratic programs that are so usable for an actual image classification task. In addition, among all the available classification methods, random forests provide one of the highest accuracies. The random forest technique can also handle big data with numerous variables running into thousands. In fact, it can automatically balance data sets when a class is more infrequent than other classes in the data that is so important for a real PolSAR image classification task. Experimental results on three well-known PolSAR data sets illustrate that the designed ESSFL can extract more discriminant features using the designed deep learning model. In the end, the experiments prove that the designed ESSFL has a significant classification performance compared with different deep learning models in the case of small number of training samples, and also it can achieve a better result in the case of large number of training samples by comparison with the most of the deep learning models. [ABSTRACT FROM AUTHOR]

The 2-D compact variational mode decomposition (2-D-C-VMD) has potential research significance in data mining of hyperspectral images (HSIs). The paper proposes a method of concatenation of features. First, the 2-D-C-VMD technique is used to decompose the HSIs to obtain smooth and clear boundary sub-band images when the 2-D-C-VMD takes the most preferred a priori parameter K ×. Second, the optimal low-frequency component features of the image and the optimal high-frequency component constructed features are concatenated first and last, and the structure of the concatenation is constructed into a new fused data. Finally, in order to avoid the redundancy of the fused data band information from affecting the classification result, combine a recently developed simple and practical superpixelwise principal component analysis (SuperPCA) dimensionality reduction method to effectively reduce the dimensionality of the fused data. By comparing the analysis with relying only on the optimal low-frequency image feature information and with the direct linear superposition fusion that has been employed, it is verified that the concatenation of features approach proposed in this paper can obtain better recognition results. [ABSTRACT FROM AUTHOR]

The recent deployment of new satellite-based hyperspectral imaging sensors, along with advances in classification methods, has enhanced the acquisition of information on the spatiotemporal patterns of Land Use/Land Cover (LULC).. With this context, the study presented in this paper seeks to assess the combined utilization of the Pixxel's technology demonstrator hyperspectral satellite mission with the Support Vector Machines (SVM) and Random Forest (RF) machine learning classifiers for LULC mapping in an agricultural area located in Queensland, Australia. Accuracy assessment of the resulting LULC maps has been done based on relevant classification accuracy metrics. It has been observed that SVMs outperformed RF in terms of both overall classification and urban vegetation cover mapping accuracy. An overall accuracy of 92.34% with kappa of 0.90 was reported for the SVMs' results, whereas for RF, the accuracy was 89.65% with 0.89 kappa, respectively. Both classifiers produced similar results for the classes of mango tree and sugarcane crops, whereas minor differences were observed in the depiction of water bodies and artificial surfaces. Our study findings provide insightful perspectives on the potential of Pixxel datasets for delineating the spatiotemporal distribution of LULC in agricultural landscapes, showcasing Pixxel's hyperspectral sensors' promising potential in this field. [ABSTRACT FROM AUTHOR]

Distributed Inverse Synthetic Aperture Radar (ISAR) systems employ multiple spatially separated radars to observe a target, with the potential to enhance radar imaging azimuth resolution and gather more target information. Due to the existence of gaps in the observation angles of each radars, the coherence between pulses of radar echo is disrupted during fusion imaging, making it challenging for the Range-Doppler (RD) algorithm to achieve well-focused images. This paper proposes an imaging method based on sparse signal recovery (SSR), using the reweighted ${l_1}$ l 1 norm with convolution and total variation as regularization terms. This approach not only yields well-focused ISAR images but also better preserves the overall target information compared to other methods, while also exhibiting excellent noise resistance. In addition, the alternating direction method of multipliers (ADMM) algorithm is utilized for the solution to reduce computational complexity. Simulation and measured data verify the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Acquiring precise change map for synthetic aperture radar (SAR) images poses challenges due to the absence of labelled datasets and the presence of speckle noise interference. This paper presents a two-stage fusion strategy coupled with capsule network classification detection method. Initially, to enhance the structure of change areas within the difference image (DI), we propose an energy ratio (ER) algorithm based on the mean ratio (MR). Subsequently, employing a two-stage fusion strategy integrating multiplication and discrete wavelet transform (DWT), we fuse the DIs generated by the three algorithms to effectively complement change information across distinct DIs. In addition, a saliency detection algorithm is introduced to mitigate noise interference and balance the number of samples. Hierarchical fuzzy c-means clustering (HFCM) is then used to pre-classify the salient fusion difference image (SFDI) and derive pseudo-labels. Finally, we apply the residual shrinkage capsule network (RSCN) to reclassify pre-classification results and obtain the change detection outcome, capable of extracting the multi-scale refinement features and adaptively learning discriminant information from changed and unchanged regions. Experimental evaluations on various SAR image datasets demonstrate the exceptional detection performance of the proposed method. [ABSTRACT FROM AUTHOR]

Domain adaptation (DA) offers an effective way to align feature distributions of the source domain (SD) and the target domain (TD) without requiring any target label samples. As a method of DA, representation learning effectively realizes the alignment of feature distributions in different domains by transferring domain knowledge. However, existing representation learning methods often focus on unilateral representation transfer, which potentially results in transfer bias. Additionally, most methods ignore the connection between domain alignment and discrimination during the DA process, which easily causes negative transfer. This paper proposes a dynamic weighted dual-driven domain adaptation (DW-D $^3$ 3 A) model that effectively addresses the aforementioned issues through bilateral feature transfer between domains and a dynamic weighted scheme. Technically, we first propose a dual-driven domain adaptation (D $^3$ 3 A) model, which employs symmetrical structures to facilitate the knowledge transfer of bilateral representations between source and target domain samples, learning the subspaces of two domains and reducing distribution discrepancies between subspaces via joint distribution-driven alignment. This process mitigates transfer bias and goes beyond previous unilateral transfer methods. Then, to alleviate strong constraints on projecting SD and TD into the same subspace in existing approaches, we apply a relaxed subspace constraint to bring the projections of SD and TD closer. Furthermore, data reconstruction is incorporated to preserve discriminant information from the original data. Lastly, we expand (D $^3$ 3 A) to DW-D $^3$ 3 A using a dynamic weighted scheme, which adjusts the weights assigned to domain alignment and discrimination based on their significance to inhibit negative transfer. Extensive experimentation on three datasets indicates that DW-D $^3$ 3 A outperforms seven other DA methods, showing its superior performance. [ABSTRACT FROM AUTHOR]

Hyperspectral unmixing (HU) has been a vital technique in hyperspectral image processing. Over the past years, many algorithms have been developed based on the nonnegative matrix factorization (NMF) framework. However, the conventional methods based on the NMF need prior information of the number of endmembers, which is usually estimated by a separate procedure. Therefore, the precision of estimating the number of endmembers will impact the subsequent unmixing effect. Thus, collaborating between the two processes is desired to avoid the propagation of errors. This paper proposes a new NMF-based method, which can estimate the number of endmembers and extract the endmembers and the abundance maps. Firstly, we introduce the weighted group sparsity regularization to adaptively eliminate the abundances corresponding to the redundant endmembers, while the abundances of the true endmembers are preserved. Then the number of endmembers can be determined by the reconstruction error function and the ${l_2}$ l 2 -norm of the abundance vectors. Secondly, since the simplex determined by the endmembers has the minimum volume among all simplexes that circumscribe the data points, we impose the total variational regularization to force the endmembers to be close to each other to minimize the simplex volume. And because the total variational regularization avoids the use of noisy pixels, it can decrease the influence of noise and improve the robustness of the solution. For the proposed model, we design an effective algorithm to solve it. Also, the experimental results based on the synthetic data and real data demonstrate that the proposed method can identify the endmembers number well and achieve robust unmixing results. [ABSTRACT FROM AUTHOR]

The current building surface reconstruction from airborne LiDAR point clouds pays more attention to the relationships between roof primitives but ignores the possible existing points of walls and the structural relationships between the roofs and walls. In this paper, we propose a novel approach that reconstructs a building surface model from airborne LiDAR point clouds based on improved structural constraints. Unlike existing hypothesis-and-selected-based methods, which impose the same constraints on all relationships between planar primitives, we design a series of strong and soft structural constraints to promote the representation of geometric and structural details of both roofs and walls in the reconstructed models. To be precise, we generate a set of candidate faces with detected planar primitives of building and vertical planes inferred from the corresponding footprint to complete possibly missing walls. Then, the potential structural representations associated with building walls in the original point clouds are identified by conflict detection between the support planes both existing points of building walls and inferred from the footprint, and on this basis, improved structural constraints are constructed for the structures consisting of different type of support planes in the set of candidate faces. Finally, a compact and watertight model is extracted from the faces of the candidate set by energy minimization with proposed improved structural constraints. Experiments on the various airborne LiDAR point clouds and aerial point clouds datasets have demonstrated that the proposed method could obtain accurate and compact models with the representation of geometric and structural details between building roofs and walls compared with other state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

In recent years, with the development of deep learning, SAR image-based ship target-detection technology has become a current research hotspot. SAR images are characterized by complex backgrounds, significant speckle noise interference, and poor interpretability. To address this challenge, this study proposes a new Gaussian circle radius determination strategy and a multi-level feature fusion codec-based ship detection network (MFFC-SDN) for SAR images. Differing from CenterNet's corner-oriented Gaussian circle strategy, the Gaussian circle strategy proposed in this paper takes the centre point coordinates of the real target as the origin to define a circular region. This approach allows for more precise supervision of the training model. Meanwhile, we propose MFFC-SDN to solve the problem of the poor performance of CenterNet in SAR ship detection. MFFC-SDN is an anchor-free, single-stage target-detection method with a low computation cost. In MFFC-SDN, we introduce a Multilevel Feature Recapture Codec Module (MFR-CM) to enhance the feature extraction capability of the network by using the features of the encoding stage twice because of the easy loss of image features in the lengthy codec process of the CenterNet network backbone. A Residual Attention Pyramid Feature Fusion Module (RA-PFFM) is designed to enhance the feature map and obtain multi-scale features. Experimental results on the SSDD and SAR-ship-dataset show that MFFC-SDN significantly enhances SAR image target-detection performance, effectively addressing detection challenges related to large target size variations and complex environmental conditions. Compared with existing algorithms, MFFC-SDN achieves the highest $A{P^{50}}$ A P 50 . [ABSTRACT FROM AUTHOR]

The precipitation estimation with high spatial and temporal resolution over Tibet is of great significance for monitoring the stability of the plateau ecosystem and early warning of natural disasters. In response to the problem of the unsatisfactory accuracy in traditional real-time precipitation inversion methods and quantitative accuracy estimation products in this region. This paper adopts deep-learning methods to explore the corresponding relationship between precipitation and multi-band observation data from geostationary satellite imagery. The U-High Resolution Network (U-HRNet) as backbone is fully exploited to generate multi-scale features and fused using attention gates to extract high-resolution spatial representations representing precipitation clouds, thereby achieving precipitation estimation of target clouds at different scales. In addition, in response to the characteristics of high precipitation intensity and drastic cloud cluster changes in summer convective weather within a short period of time, a multi-head attention mechanism is introduced to extract temporal information from time series inputs and fuse it with spatial features to improve the accuracy of real-time precipitation estimation. Compared with the operational precipitation products and baseline deep-learning models, the classification metrics such as Probability of detection (POD), False Alarm Ratio (FAR) and Critical success index (CSI) show that the proposed model (U-HRNet+) has better performance in test set and has high generalization ability for estimating short-term heavy rainfall over Tibet and its surrounding areas in summer. Therefore, the proposed model has the potential to serve as a more accurate and reliable satellite-based precipitation estimation product. [ABSTRACT FROM AUTHOR]

Spectral clustering, as an algorithm based on graph theory and spectral theory, has shown excellent performance in the classification tasks of hyperspectral images in recent years. Although better results have been achieved, some challenges still exist. The inclusion of a priori information can increase the performance of spectral clustering algorithms; however, in practice, it is often unable to meet the demand for a priori information; at the same time, spectral clustering for the quality of the similarity matrix is demanding, and some of the current algorithms are to improve the similarity matrix from the aspect of data planning, and ultimately for the label feature matrices are diluted. In response to the above problems, this paper proposes a self-supervised spectral clustering with spectral embedding (SESSC). The algorithm obtains a low-dimensional representation of the data through spectral embedding, which can simplify clustering while preserving feature information; then uses the similarity matrix about the data and the guidance of the sample constraint information to obtain a new similarity matrix in order to further refine the structural graph, and the result can feed back and optimize the label feature matrix and the low-dimensional representation of the data. Additionally, we introduce a fractional theory in the update of the sample variable matrix, which assures the integrity and validity of the information in the update. Experimental results show that the proposed algorithm has better performance in hyperspectral image classification than existing spectral clustering algorithms. [ABSTRACT FROM AUTHOR]

There are challenges for cloud detection over bright surfaces, especially for Sentinel-2 image without thermal band. In this paper, we developed a new cloud detection algorithm for Sentinel-2 data designed to tackle the challenges of cloud detection on bright targets. Our algorithm (called RF-NDTeI-SNIC) was built through the integration of Random Forest (RF) model, Normalized Difference Temporal Index (NDTeI) cloud index and Simple Non-Iterative Clustering (SNIC) image segmentation algorithm. First, RF and NDTeI were used to derive the initial and second stage's cloud detection results, respectively. Then, the RF- and NDTeI-based results were fused to remove the bright surfaces and land cover changes. Finally, the fusion results were further refined using SNIC for morphological processing. The proposed RF-NDTeI-SNIC algorithm was evaluated using nine challenging cloud-covered Sentinel-2 scenes with substantial bright surfaces. Results indicated that the average overall accuracy of our algorithm was 95.01%, with average commission rate of 10.19% for cloud. In addition, five commonly used Sentinel-2 cloud mask algorithms including s2cloudless, QA, Fmask, CDI and Tmask were selected for comparative analysis. Results suggested that our algorithm outperformed others with overall accuracies of 89.60–94.17% and cloud commission rate of 9.46–37.50%, and have significant advantages in terms of bright surfaces. In summary, the RF-NDTeI-SNIC algorithm we developed was capable of yielding accurate Sentinel-2 cloud masks, and the novel NDTeI cloud index we proposed gave a new and effective approach for improving the separability of clouds and bright surfaces. [ABSTRACT FROM AUTHOR]

The fusion of hyperspectral and multispectral images presents a challenge as it involves blending a low-resolution hyperspectral image (HSI) with a corresponding multispectral image (MSI) to produce a high-resolution hyperspectral image (HRI). A number of existing techniques have limitations; for instance, matrix decomposition-based approaches fail to retain adequate spatial and spectral image information during fusion, while tensor decomposition-based processes have high computational overhead. In this paper, we propose a novel method for fusing hyperspectral and multispectral images. Our method leverages the strong correlation among the spectral bands of hyperspectral images and employs the SVD technique to extract spectral feature subspaces. This approach results in a more compact and representative feature space for fusion. Secondly, the proposed method utilizes Elastic Net regularization in combination with ${L_1}$ L 1 and ${L_2}$ L 2 regularization for effective feature selection of highly covariate features. Weighted group sparse regularization is employed to enhance the fusion effect, enabling better representation of the image's structure and features. The algorithm is subsequently evaluated on multiple datasets to confirm its effectiveness. The results of the experiment suggest that the suggested algorithm greatly enhances the spatial resolution and visual clarity of HSI images while preserving the spectral characteristics when compared to conventional methods for fusion of hyperspectral images. Additionally, the constraint for regulation can competently reduce any noise or artefacts, thereby boosting image discrimination. To summarize, the utilization of a sparse tensor-based hyperspectral image fusion algorithm with subspace learning provides an efficacious approach for processing hyperspectral imagery. This method is capable of improving spatial resolution, extracting advantageous features from hyperspectral imagery, and ultimately supports the process of remote sensing image analysis and application. [ABSTRACT FROM AUTHOR]

Hyperspectral images are easily contaminated by multi-modal noise in the process of data acquisition. Generally, hyperspectral signatures and noise appear unexpectedly scattered in the spatial-spectral domain. Conversely, in the transform domain, the principle components of an image are gathered in the low-frequency bands, while the details and noise often occur in high-frequency parts. The existing denoising methods usually divide the whole frequency domain into two parts, that is, high frequency and low frequency, which can not make the most of the knowledge contained at different frequencies or scales, especially in the multiscale transform domain. Furthermore, aiming to restore a clean image from the noise-corrupted data, hyperspectral image denoising often amounts to an ill-posed problem, in which reshaping cubic data into matrix form results in the loss of structure information. To address the aforementioned issues, this paper incorporates the reweighted smoothing regularization in the transform domain into the noise-perturbed degradation model to reformulate a novel optimization problem. First, the multiscale curvelet is leveraged to transform the spatial matrix into frequency domain, where the coefficients are carefully designed to weight the components at different scales and orientations. Second, ${l_1}$ l 1 -norm is imposed on the abundance term to promote sparsity in the transform domain for performance enhancement. Finally, using proximal alternating optimization, an efficient algorithm is well developed to obtain the closed-form solutions to the proposed denoising problem. The experimental results on several synthetic and real datasets, illustrate that the proposed method improves PSNR by at least 1.5 dB and reduces ERGAS error by more than 35% than the state-of-the-art approaches under the multi-modal noise mixed environments, which verifies the validity of regularization terms in hyperspectral images denoising. [ABSTRACT FROM AUTHOR]

Radar signatures of rain over the ocean have a complex structure since they receive contributions from surface scattering and volume scattering and attenuation by hydrometeors in the atmosphere. These contributions overlap and are often difficult to detangle. While most of the mechanisms contributing to radar signatures of rain over the ocean are well understood, there is one remaining issue that has been discussed controversially in the literature for a long time. It is the question what scattering mechanism causes the areas of strongly enhanced radar backscatter, also called 'bright blobs' or 'bright patches', which are frequently observed on spaceborne C-band SAR images acquired over tropical and subtropical oceans in the presence of convective rain. Recently, papers have been published in which it is hypothesized that they are caused by radar backscattering at hydrometeors in the melting layer (ML). Although many observational facts seem to support this hypothesis, there exists one strong argument against this hypothesis: It is the observation that the position of the ML radar signatures (bright blob) in the SAR image is not shifted in anti-range from the position, where the rain column hits the sea surface. This absence of a shift is observed when 1) comparing Sentinel-1 SAR images on which rain cells are visible with quasi-concurrently acquired weather radar images and 2) when inter-comparing of SAR images of rain cells acquired concurrently at different frequencies and polarizations. Based on these observations, we discard the hypothesis that the bright blobs are due to volume scattering at hydrometeors in the ML and hypothesize instead that they are due to scattering at splash products at the sea surface. This hypothesis is supported by radar backscattering measurements carried out in the laboratory and from a shore-based platform, which show that, at C-and X-band, strong rain can give rise to strong radar returns also at cross-polarization. [ABSTRACT FROM AUTHOR]

In this paper, sea surface waves during complex sea state (i.e. tropical cyclones (TCs) and Kuroshio current) are inverted from dual-polarized (vertical – vertical (VV) and vertical – horizontal (VH)) synthetic aperture radar (SAR) images using an existing algorithm, namely, the parameterized first-guess spectrum method (PFSM). In particular, the wind term is included in two modulation transfer functions (MTFs), i.e. tilt modulation and velocity bunching. More than 3000 Sentinel (S-1) images acquired in interferometric wide (IW) mode and extra wide (EW) mode are collected, and the operational CyclObs wind product provided by the French Research Institute for Exploitation of the Oceans (IFREMER) is collected corresponding to the TC images. Note that these images are located within 500 km away from TC eyes. The wave spectra during the TCs are then inverted using the PFSM algorithm and the SAR-derived wind product. Simultaneously, the 0.05° gridded wave spectra collocated with these images are simulated using a third-generation numeric wave model, called WAVEWATCH-III (WW3), in which the reconstructed wind based on the European Centre for Medium-Range Weather Forecasts (ECMWF), the Copernicus Marine Environment Monitoring Service (CMEMS) current, and the CMEMS sea level are treated as the forcing field. Comparison of the SAR wave retrievals and simulations reveals that the root mean square errors (RMSEs) of the significant wave height (SWH) and mean wave period (MWP) are 0.55 m and 0.43 s, their correlation coefficients (COR) are 0.94 and 0.92, and their scatter index (SI) values are 0.17 and 0.07. These results demonstrate the better performance of the PFSM algorithm using theoretical MTFs, i.e. an RMSE of 0.68 m and COR of 0.90 for the SWH and an RMSE of 0.60 s and a COR of 0.87 for the MWP. Although the accuracy of the wave retrieval from the SAR image under complex sea state is improved when implementing the tilt and velocity bunching MTFs with the strong wind term, SWH retrievals by algorithm PFSM suffer saturation problem at SWH > 8 m. [ABSTRACT FROM AUTHOR]

Land-cover classification based on remote sensing technology has been adopted for decision-making concerning agricultural development, urban planning, and ecosystem protection in arid oasis zones. The semantic segmentation method based on deep learning, as a new paradigm, can effectively overcome the limitations of traditional pixel-based and object-based methods and obtain good classification results from high spatial resolution (HSR) remote sensing images. However, how to extract the exact category boundary and realize the high precision mapping is still a problem. This paper proposes a novel high-resolution feature pyramid attention network (HRFPANet) for land-cover classification. It effectively integrates the advantages of multi-scale feature extraction, attention mechanism, and feature fusion and alleviates boundary inconsistency, roughness, and category fragmentation associated with previous semantic segmentation models. The experimental results show that the mIoU score of HRFPANet is 79.5%, which is 11.5% and 2.6% higher than that of PSPNet and UPerNet, respectively. It proves the proposed model can be used for qualified land-cover mapping in arid oasis zones. Our source code is available at . [ABSTRACT FROM AUTHOR]

The intelligent segmentation of high-resolution remote sensing (HRS) image, also called as dense prediction task for HRS image, has been and will continue to be important research in the remote sensing community. In recent years, the growing wave of artificial intelligence (AI) technology has introduced innovative paradigms to this domain, yielding outstanding results and overcoming many challenges with conventional segmentation techniques. This paper provides a comprehensive review of these intelligent segmentation methodologies, including traditional pattern recognition, convolution neural network (CNN)-based, and Transformer-based techniques. However, the explosive but incomplete development of intelligent segmentation techniques also poses more challenges for earth observation experts, the most of which is the technical interpretability. Consequently, we consider these segmentation techniques in the aspect of explainable artificial intelligence (XAI). Data-centric XAI thinks the practical applications of the segmentation model while model-centric XAI will facilitate the understanding of decision-making processes and the adjustment of structural features. Moreover, this review identifies novel research questions and provides constructive insights and recommendations to HRS image segmentation tasks, which may shed new light on the intelligent segmentation methods within the remote sensing image understanding community. [ABSTRACT FROM AUTHOR]

Marine mucilage, also known as sea-snot, is an environmental disaster with ecological, economical, and public health-related impacts, which is occurring with increasing frequency and severity worldwide. One of the latest mucilage occurrences, the mucilage outbreak of Spring 2021 in the Sea of Marmara, lasted for well over three months and covered more than area of 1160 $k{m^2}$ k m 2 , and was repeated in 2022, although to a smaller extent in terms of duration, coverage area, and severity. Remotely sensed images acquired by Earth observation sensors provide a convenient and promising alternative for environmental monitoring of marine mucilage, with respect to field and laboratory work which are costly, require expert knowledge, and which have to be conducted in a small time-window and have limited generalizability due to the under-sampling of a large medium with a highly dynamic character. This work presents an unmixing-based approach using nonnegative matrix factorization for marine mucilage monitoring from remotely sensed hyperspectral data. The proposed method enables to identify the spectral signature of mucilage, and its variants where applicable, and the precise detection and estimation of mucilage concentrations in affected regions. Additionally, the proposed approach facilitates monitoring the temporal variations of mucilage presence, distribution, and accumulation, between data acquired from the same region at different dates or times. This enables quantitative spatio-temporal mapping of distributions and the potential of assessing the severity of outbreaks, without the requirement of manually prepared ground truth maps or carefully selected thresholds. Real hyperspectral data acquired by the PRISMA mission are used to validate the proposed approach qualitatively and quantitatively. The paper also discusses future directions and recommendations to enhance the performance and applicability of the proposed approach for marine mucilage analysis, and environmental monitoring in general. [ABSTRACT FROM AUTHOR]

The use of deep learning-based high resolution remote sensing image sea and land segmentation method has become prevalent in various fields such as environmental monitoring, resource assessment, and urban planning. However, the segmentation process faces challenges due to the complex coastline and the similarity in spectral colour difference between the sea and land in certain areas. The remote sensing images obtained from satellites often exhibit unclear boundaries between the sea and land, as well as intricate coastlines. Consequently, the segmentation of sea and land becomes a difficult task. The existing sea-land segmentation methods have limitations in accurately segmenting long and narrow waters, as well as areas with low differentiation between sea and land. In this paper, we propose a new sea-land segmentation method called SRMA (Swin-Res-Multi-Attention). SRMA performs parallel feature extraction using a dual-branch parallel: Swin-Transformer branch and ResNet branch. The combination of Swin-Transformer's self-attention and ResNet's residual design helps reduce the loss of feature information, thereby avoiding incorrect segmentation in regions with low distinctness. To improve overall stability, we introduce the MCA (Multiscale channel attention module) module, which combines feature information from different scales. The multi-scale pyramid design of MCA module ensures stability and improves segmentation accuracy in narrow waters. Experimental results on a public dataset demonstrate that our model outperforms existing methods. Additionally, we evaluate the model's performance in complex scenarios by creating a new dataset that includes a complex urban water body and cloud interference. The results showcase the robustness and superiority of our model. [ABSTRACT FROM AUTHOR]

This paper addresses the challenge of training deep learning models for fire smoke scene detection from multi-sensor, multi-spectral satellite imagery, where spectral bands vary and training data is scarce, especially for new sensors. We introduce a novel cross-sensor transfer learning approach enhanced by spectral pattern extraction, notably aided by a deep learning (DL) module called Input Amplification (IA). Our contribution includes the creation of two datasets, Landsat_smk2770 and Sentinel2_smk, with varying spectral bands. We employ a model named IA_VIB_SD which incorporates IA and is designed specifically for fire smoke scene detection. We investigated the transferability of the Landsat_smk2770-trained IA_VIB_SD model to the Sentinel2_smk dataset using various transfer learning techniques, and compared the performance of the transferred model with the benchmark IA_VIB_SD model exclusively trained on pure Sentinel2_smk data. Our proposed transfer learning approach resulted in a transferred model with an average accuracy 5% higher than the benchmark model. Notably, Our proposed transfer learning approach outperformed existing transfer learning methods by more than 1% in terms of the accuracy, even when trained on a mere 10% of the Sentinel2_smk dataset. This research advances fire smoke scene detection from multi-sensor, multi-spectral satellite imagery. [ABSTRACT FROM AUTHOR]

In this paper, an improving spatiotemporal image fusion (STIF) incorporating unmixing step by considering the point spread function (PSF) effect model (CPSF) is proposed to address the problem that unmixing accuracy in STIF models incorporating unmixing step is vulnerable to the PSF effect. First, an unsupervised clustering algorithm is used to cluster the fine image at previous date into a land cover map, and the coarse class fraction matrix is generated from this land cover map. Second, the fine class fraction matrix uncontaminated by the PSF effect is calculated by upsampling the coarse class fraction matrix using the Area-to-Point Kriging (ATPK) method under the PSF constraint. Third, this fine class fraction matrix is convolved with the ideal square wave filter to downsample, and then normalized to obtain the coarse class fraction matrix without the influence of the PSF effect. Finally, the CPSF-processed class fraction matrix is used to replace the original one in the following unmixing and prediction process to generate the fine resolution image at the prediction date. Experiments on two remote sensing data sets show that the proposed method can effectively improve the accuracy of spatiotemporal image fusion methods incorporating unmixing step. [ABSTRACT FROM AUTHOR]

The development of the modern urban economy is closely tied to road construction, with roads being one of the most crucial components in urban development. The use of high-resolution remote sensing images to monitor the road conditions in cities and surrounding towns has become a highly emphasized research focus in recent years. However, due to the complexity of the environment, this task still faces significant challenges. For instance, urban road structures are intricate, which involve multiple lanes, intersections, building obstructions, etc. Rural roads may consist of narrow paths or irregular dirt roads. In addressing these issues, this paper proposes a road extraction algorithm based on a Dual-Encoder-Decoder U- Net (DEDU-Net). The algorithm leverages a dual encoder-decoder network to extract multi-scale information from the image. It then employs a dual decoder network to restore the feature map to the original image, achieving precise road extraction. Additionally, a new module, the Global Fusion Module (GFM), is introduced. This module achieves global context information fusion by weighting features. In the experimental section, two publicly available datasets were used for testing: the CHN6-CUG dataset and the Gansu Mountain Road dataset. For example, for the Gansu Mountain Road and CHN6-CUG mixed dataset, the model achieved an Overall Accuracy (OA) of 94.947% and a mean Intersection over Union (mIoU) of 70.971%. The results indicate that compared to traditional methods, this proposed method exhibits higher accuracy and robustness. It can adapt to both urban and rural roads, delivering outstanding performance even in complex scenarios. [ABSTRACT FROM AUTHOR]

Remote sensing detection is a difficult task that requires not only identifying objects with complex background, quality, and angle issues, but also being lightweight enough to be carried by edge devices. In real remote sensing scenarios, achieving accurate, fast, and low-resource-consumption automated detection remains a significant challenge. Therefore, this paper proposes an efficient balanced network (EB-Net) for real remote sensing devices. First, a dynamic sparse attention (DSA) mechanism is proposed and has been proven with high performance via the complexity analysis. In addition, a new dynamic sparse transformer (DSFormer) is constructed using DSA, which enhances feature information and adapts to image resolution by self-attention and multi-headed attention, and achieves more flexible computation by random sampling. Then, three versions of discrete distribution IoU (DDIoU) are defined for adapting various scenarios and tasks in remote sensing, and this loss function makes the model achieve high accuracy and more lightweight. Finally, to make the model more lightweight, a cropped SPPF (CroSPPF) is presented, which significantly improves the computational efficiency by the lightweight of the sequence structure and activation function. Ablation experiments are conducted on the NWPU VHR-10 dataset, and demonstrate the effectiveness of the proposed methods. Numerous comparisons with state-of-the-art detectors are conducted on the NWPU VHR-10, RSOD and DOTA datasets. The experiments show that EB-Net outperforms state-of-the-art remote sensing detection models in comprehensive performance and achieves end-to-end accurate and lightweight detection. [ABSTRACT FROM AUTHOR]

Multiple kernel learning (MKL) is an efficient way to improve hyperspectral image classification with few training samples by integrating spectral and spatial features. Nonetheless, presenting a MKL algorithm to capture significant information and eliminate redundancy is indispensable. In this paper, a novel graph-based MKL method (GMKL) is introduced to enhance the performance of the MKL algorithm and achieve high discriminative features, resulting in maximum separability. To do so, an informative low-rank graph is determined via a new algorithm called block diagonal representation that takes into account both local and global data structures. Then, in lieu of a time-consuming search for finding the optimal combination of the basic kernels, unlike the prior works, an optimal projective direction is acquired using graph theory. Various basic kernels generated from superpixels and attribute profiles are projected to produce a discriminative combination of kernels. Eventually, the optimum kernel is given to the support vector machine classifier. Experiments have been conducted on two popular hyperspectral datasets (Indian Pines and Pavia University). For the datasets, the proposed algorithm gained 90.43% and 99.28% overall accuracy, respectively, using only 1% training samples, which is much better than the state-of-the-art MKL algorithm and deep learning-based approaches with lower computational time. [ABSTRACT FROM AUTHOR]

L-band spaceborne polarimetric synthetic aperture radar (PolSAR) data has been well used for the high-resolution retrieval of the ionospheric Faraday rotation (FR). A vast majority of these retrievals were, however, conducted on PolSAR data of lands, with only a very few over oceans. For the global imaging of the Earth's ionosphere, the retrieval of oceanic ionosphere is indispensable. This paper aims to furnish a deep insight into the FR retrieval on oceans using the widely used Bickel-Bates estimator based on the two cross-circular polarizations ${Z_{12}}$ Z 12 and ${Z_{21}}$ Z 21 . The FR estimates (FRE) on oceans are evaluated in detail by comparing with those on lands on four ALOS PALSAR datasets. The achieved oceanic FRE are shown to possess comparable or even better quality than those on lands. It occurs in three of the four scenes that the selected ocean sub-area holds a smaller standard deviation (STD) of FRE (less than 0.7000°) than all the selected land sub-areas. Especially, the quality of FRE is shown to be closely linked to the dispersed degree of the magnitude of the conjugate product of ${Z_{12}}$ Z 12 and ${Z_{21}}$ Z 21 via the analysis of coefficient of variation (CV), according to which the unusual occurrence of larger magnitude level corresponding to larger STD of FRE can be explained. Nevertheless, compared with CV, the magnitude of the polarimetric coherence between ${Z_{12}}$ Z 12 and ${Z_{21}}$ Z 21 (i.e. $\left| \gamma \right|$ γ ) is identified as a better index to measure the quality of FRE. A smaller STD of FRE appears along with a larger $\left| \gamma \right|$ γ . Three of the four scenes show that the selected ocean sub-area bears a larger mean of $\left| \gamma \right|$ γ (greater than 0.9890) in comparison with all the selected land sub-areas. This work can promote the ionospheric research over oceans and to the correction of ionospheric effects in oceanic L-band spaceborne polarimetric radar data as well. [ABSTRACT FROM AUTHOR]

Hypergraph neural networks (HGNNs), extending the techniques of graph neural networks, have been applied to various fields due to their ability to capture more complex high-order node relationships. However, for hyperspectral image (HSI) classification tasks, previous HGNN-based works usually constructed hypergraphs using pixels as nodes, resulting in massive computational costs. Meanwhile, pixel-level personalized features are required for HSI classification. To achieve high efficiency and accuracy simultaneously, this paper presents a fast hypergraph neural network with detail preservation (DPFHNet) for HSI classification. It constructs hypergraphs at the superpixel level to reduce time consumption and supplement pixel-level detail features through a classification refinement module. This framework contains multiple stages. Firstly, its main stage is designed with HGNNs from a superpixel viewpoint rather than pixels, providing a fast strategy to capture high-order complex relationships of multiple homogeneous irregular regions. After that, auxiliary stages based on convolutional neural networks are integrated into the main stage, which adopts a hierarchical design and attempts to acquire pixel-level spatial-spectral information before the hypergraph feature extraction of the main stage, assisting in learning more valuable features. Finally, a classification refinement module is constructed, which generates pixel-level detail features to refine the superpixel-level features obtained by HGNN. Experiments on three datasets illustrate that DPFHNet achieves competitive results and efficiency compared to advanced methodologies. [ABSTRACT FROM AUTHOR]

Advances in hardware capabilities and big data technologies over the past decade have enabled the application of deep learning techniques to address the challenge of super-resolution in remote-sensing images. Recently, while deep-learning-based methods have outperformed traditional methods, the abundance of information in remote-sensing images creates an imbalance between performance and computational resource consumption in current deep-learning-based methods. This paper introduces a stable super-resolution algorithm based on parameter reconstructive optimization to address these issues. First, based on the stable super-resolution generative adversarial network (SSRGAN), the algorithm employs a generator with networks using residual connections to reconstruct images with enhanced resolution. Next, it extracts content, adversarial, and regularization losses using the discriminator from a stable super-resolution generative adversarial network, which in turn guides the training of the network. Finally, upon completion of training, structural re-parameterization is conducted to optimize the multi-branch trained network into a plain inference network. This inference network can serve as an individual new generator. The results of qualitative and quantitative experimental comparisons with the models Bicubic, super-resolution convolutional neural network, very deep super-resolution convolutional network, deep recursive residual network, super-resolution generative adversarial network (SRGAN), enhanced SRGAN, and stable SRGAN on the Gaofen-5 AHSI satellite dataset suggest that this algorithm achieves improved evaluation indices with a 4× magnification ratio, reaching a peak signal-to-noise ratio of 30.7207 dB and structural similarity index measure of 0.8178. Compared with the trained but unconverted generator, which can also work independently, implementing re-parameterization results in approximately a 10% reduction in the number of parameters, indicating lower resource consumption, while the reconstruction effect is minimally influenced. Furthermore, the super-resolution results exhibit richer detail, increased contrast, and better scene adaptability. [ABSTRACT FROM AUTHOR]

Multispectral image (MSI) and hyperspectral image (HSI) fusion is a popular method for HSI super-resolution reconstruction (HSI-SR). MSI-HSI fusion problem is ill-posed and demands several image priors or regularization terms to solve accurately, which is a challenging issue. In this paper, we propose an MSI-HSI fusion model via subspace representation and nonlocal low-rank regularization (SRNLRR). SRNLRR model incorporates the global spectral correlations and spatial nonlocal similarities of HSI to improve the fusion results, where the priors complement each other. First, we use the mode-n tensor-matrix product to project latent high spatial resolution HSI (HR-HSI) into spectral subspace, which can capture spectral low-rankness and reduce computational complexity. Then, based on low-rank representation (LRR) and nonlocal processing strategy, we design a spatial nonlocal LRR regularization (spa-NLRR) and a spectral global LRR regularization (spe-GLRR). These two regularizations analyze the spatial nonlocal similarities and spectral global correlations from intermediate-level vision. Finally, we use the residual regularization program to obtain more image information and input it into the fusion model. We use alternate minimization (AM) methods to optimize the SRNLRR model and employ the alternating direction method (ADM) on spatial/spectral LRR learning. Comparison experiments between the SRNLRR method and six advanced methods on three HSI datasets indicate the superiority of our method. [ABSTRACT FROM AUTHOR]

In inverse synthetic aperture radar (ISAR) imaging, the conventional range-Doppler (RD) algorithm cannot obtain satisfactory imaging results for sparse aperture. Compressed sensing (CS) imaging methods, such as the typical iterative shrinkage and thresholding algorithm (ISTA), are often used in sparse aperture radar imaging. However, CS-based methods have high computational complexity and difficulty in setting parameters. To overcome the shortcomings, a new deep unfolding network named complex-valued weighted learning ISTA (CV-WLISTA) is proposed in this paper. The new network is very efficient and can learn the parameters adaptively. Based on ISTA and a new weighting strategy, it takes advantage of both the sparsity and the structural property of ISAR images to improve imaging performance. The experimental results show that CV-WLISTA has better reconstruction performance and higher efficiency compared with the traditional algorithms. Therefore, CV-WLISTA is an efficient ISAR imaging method with excellent performance. [ABSTRACT FROM AUTHOR]

Automatic building detection from remote sensing images holds significant applications in various domains, including cartography, land-use change detection, urban planning, and 3D intelligent city construction. Driven by the innovation of deep learning technology, convolutional neural networks (CNNs) have emerged as a powerful framework for automatic interpretation, exhibiting hierarchical feature generation and robust representation for building extraction. However, the inherent computational limitations of convolutional operators hinder the acquisition of an effective global receptive field. Recent advancements in deep learning have witnessed the adoption of the multi-head self-attention (MHSA) mechanism by Transformer, presenting the superiority of global context construction. Nonetheless, this network structure encounters difficulties in effectively modelling two-dimensional spatial structure features with scale variation in semantic segmentation and often has huge parameters with high algorithm complexity. To address the above problems, this paper proposed a hybrid deep neural network synergistically harnessing the superiority of CNN and Transformer to extract buildings from remote sensing images. Meanwhile, an improved MHSA mechanism is developed to facilitate the construction of multiscale global representations, concurrently reducing computational complexity. Furthermore, the cross-level MHSA modules the semantic correlation between the deep and shallow-level features to refine spatial fine-grained information. The proposed method was comprehensively evaluated on two building datasets via quantitative analysis and visualization. The experimental results confirmed that the developed network and modules outperform other state-of-the-art methods and can effectively improve the accuracy (93.4% F1-score on the WHU dataset, 89.38% F1-score on the CAMB dataset) with high efficiency for building extraction. [ABSTRACT FROM AUTHOR]

Multilabel remote sensing image classification (MLRSIC) has received increasing research interest. Taking the co-occurrence relationship of multiple labels as additional information helps to improve the overall performance. However, current methods only focus on using it to constrain the final feature which is output from a convolutional neural network (CNN). On the one hand, these methods need to exploit the potential of label correlation in feature representation fully. On the other hand, they increase the label noise sensitivity of the system, resulting in poor robustness. In this paper, a novel method called 'Semantic Interleaving Global chaNnel Attention' (SIGNA) is proposed for MLRSIC. First, the label co-occurrence graph is obtained according to the statistical information of the training set and fed into a graph neural network (GNN) to generate optimal semantic feature representations of each label. Next, the semantic features are interleaved with visual features which are extracted by CNNs to guide the overall features of the input image transform from the original feature space to the semantic feature space with embedded label relations. Then, global attention triggered by semantic interleaving is used to emphasize visual features in important channels. Finally, to make SIGNA easier to use and more optimized, multihead SIGNA-based feature adaptive weighting networks are proposed as plug-in blocks to plug into any layers of a CNN. For remote sensing images, better classification performance can be achieved by inserting the plug-in blocks into the shallow layers of CNNs. We conducted extensive experimental comparisons on three data sets: UCM, AID and DFC15. Experimental results demonstrate that the proposed SIGNA achieves superior classification performance compared to state-of-the-art (SOTA) methods. Notes that the codes of this paper will be open to the community for reproducibility research. [ABSTRACT FROM AUTHOR]

Building roofs are crucial elements that must be extracted from satellite images for use in services like updating geodatabases, risk analysis and rescue maps. The diversity and complexity of the building structures present challenges for accurate and efficient building rooftop extraction from very high resolution (VHR) images. The significance of building extraction from satellite imagery in various applications has attracted the attention of the research community in this domain. In this paper, the progression of the research in building extraction is presented based on two streams, such as Traditional building extraction techniques and Evolving DL-based building extraction techniques. Furthermore, the categorization of traditional techniques into transform-based techniques, indices-based methodologies and machine learning-based methodologies are summarized. An enhanced comprehension of the challenges associated with extracting buildings from aerial images using DL models is given. A performance analysis of the popular Deep Learning models used for building extraction, such as UNet, SegNet, ResUNet, UNet++, ENRUNet and Dilated ResUNet, in terms of Accuracy, IoU and Dice Loss is presented. We hope that this paper will improve researchers' understanding of the problems associated with extraction of buildings from images and ultimately to develop new models with better performance. [ABSTRACT FROM AUTHOR]

Remote sensing scene classification is the basis of advanced smart urban planning tasks such as urban functional zone division and land use type identification. In recent years, a wide range of emerging data sources is receiving progressive attention for urban features extraction, such as satellites, unmanned aerial vehicles (UAVs), and ground sensors. How to effectively utilize these multi-view data jointly to improve scene classification performance has become a hot topic of remote sensing challenge. Existing feature fusion methods tend to map data from different views into a common feature space, which is often difficult to find when the data between views differ greatly. Furthermore, because these methods require data from all views as input, they are not flexible enough to handle situations where there is only one view input when inferring. To address the aforementioned issues, a novel Coupled Parallel Architecture (CPA) using Weighted Collaboration Fusion Constrained by Consistency Between Views (CBV-WCF) is proposed in this paper. In the training phase, the CBV module reduces the impact of the heterogeneous gap across views by capturing the consistency information between views. Well the WCF module is used to fully mine and effectively fuse the complementary information between views to improve the performance of downstream tasks. In the inference phase, the proposed architecture can effectively improve the classification performance in both cases of multi-view and single-view input. Our method is evaluated on air-ground dual-view scene classification, which is a typical multi-view task with large image differences between views. Experimental results on two publicly available air-ground dual-view datasets demonstrate that the proposed framework significantly improves classification performance while bringing some inspiration and new solutions to multi-view tasks. The code of this paper will be published at: . [ABSTRACT FROM AUTHOR]

Remote sensing image scene classification is a fundamental task in intelligent interpretation of remote sensing images. Although Transformers possess a powerful attention mechanism, they require lengthy training procedures to achieve good performance levels. To address this issue, this paper proposes a novel deep learning network model by combining CNN and Swin Transformer named SwinHCST. Firstly, the model uses Weighted Normalized CNN to quickly extract low-level features of the image. Secondly, the Receptive Field Block module facilitates multi-scale information fusion, Thirdly, the Information Fusion Transformer further excavates the deep-level features of the image. Furthermore, this paper has designed a plug-and-play Cross Spatial Information Fusion Block, which is used to encodes dimensional information and extracts global information to enhance information exchange. The scene classification experiments show that the proposed model outperforms other methods on the three selected datasets and can achieve excellent performance without requiring large amounts of data and training. Specifically, the classification accuracy of the proposed method on the three datasets is 93.76%, 93.60%, and 98.10%, which is 1.7% to 3.71% higher than ResNet50 and 3.7% to 5.7% higher than Swin Transformer. [ABSTRACT FROM AUTHOR]

Semantic segmentation of high-resolution remote sensing images is important in land cover classification, road extraction, building extraction, water extraction, etc. However, high-resolution remote-sensing images have a lot of details. Due to the fixed receptive field of convolution blocks, it is impossible to model the correlation of global features. In addition, complex fusion methods cannot integrate spatial and global context information. In order to solve these problems, this paper proposes a cross-linear attention network (CLANet) to capture spatial and context information in images. The structure consists of a spatial branch and a context branch. The spatial branch is constructed by stacked convolution to better capture spatial information. The context branch models the global information based on the transformer deformation module. In addition, to effectively fuse spatial and context information, this paper also designs a feature fusion module (FFM), which uses a cross-linear attention mechanism for feature aggregation. Finally, this paper conducts many experiments on the ISPRS Vaihingen and the ISPRS Potsdam datasets. Among them, 82.28% of mIoU achieves on the ISPRS Vaihingen dataset. The experimental results show that CLANet has better performance and effect than the methods in recent years. [ABSTRACT FROM AUTHOR]

Semantic segmentation is a fundamental but meaningful task in the remote sensing image understanding community. Great progress has been made in optical sensor photography technology, which poses an opportunity and a challenge for remote sensing image segmentation task. But, in fact, a longstanding and intractable problem is that many hard pixels in special position, i.e. the prevalent intra-class noise and a poor boundary delineation, is difficult to classify due to their inherent uncertainty. In this paper, we comprehensively consider the characteristics of deep learning and introduce traditional pattern recognition methods to drive structure learning, which can leverage the corresponding fuzzy logic model to alleviate the aforementioned problem in remote sensing images. Specifically, this paper designs a multiscale bidirectional fuzzy-driven learning network (MBFNet), which takes advantage of both deep learning and fuzzy logic to effectively alleviate the inherent uncertainty of these hard pixels. The structure of convolutional neural networks driven by fuzzy systems also provides a new modelling paradigm for solving the uncertain problem in remote sensing images. Meanwhile, multiscale techniques and bidirectional fusion are introduced to enhance feature aggregation and avoid the potential adverse effects of fuzzy systems, respectively. Experimental results on two datasets demonstrate qualitatively and quantitatively that the proposed MBFNet is competitive. [ABSTRACT FROM AUTHOR]

As a classic problem of high-resolution remote sensing image tasks, land cover classification has many challenges, for instance, variable scales and low-edge discrimination. Traditional spatial semantic segmentation methods based on single-scale feature extraction are challenging to model multi-scale features. They cannot effectively fit high-resolution remote sensing images with variable scales and are prone to misclassification issues. On the other hand, spatial multi-scale semantic segmentation methods lack edge constraints, which can easily lead to edge segmentation ambiguity issues. We propose a segmentation network based on a multi-scale frequency-domain attention gating mechanism (MFAGNet) to solve the above problems. Specifically, we obtain multi-scale features through multi-scale input and design the discrete cosine transform channel attention module (DCTCAM). In DCTCAM, we extract global low-frequency features and effectively alleviate misclassification caused by scale changes. The edge features of the image are obtained through the high-frequency feature extraction module (HFEM) by using edge detection operators to enhance the model's ability to recognize edges. Additionally, no further pooling operations are employed except for a single max pooling layer in the backbone network. In this paper, we design a new gating mechanism to dynamically control the extraction ratio of various frequency features so that it can adapt to input images of different scales. This paper conducts comparative experiments on three high-resolution land cover datasets: GID, Vaihingen, and Potsdam. The new algorithm proposed compares with classic semantic segmentation models such as DANet, PSPNet, SegNet, UNet, DeepLabv3+, FcaNet, BiSeNet, AFN, and FastFCN. The experimental results show that MFAGNet is superior to the comparison models. With no significant increase in computing overhead, MFAGNet obtains MIoU of 69.05%, 73.40%, and 79.42% on the standard data sets GID, Vaihingen, and Potsdam, respectively. [ABSTRACT FROM AUTHOR]

For spaceborne ISAR imaging of space targets, due to the high-speed relative motion between the radar and targets, both the translational and rotational motion exhibit the high-order characteristics. Conventional translational motion compensation (TMC) methods are ineffective in handling such high-order translational motion. Moreover, the migration through resolution cell (MTRC) stems from the presence of the high-order spatial variant phase of the echo signal, resulting in severe defocusing of ISAR images. In this paper, a novel spaceborne ISAR imaging approach for space targets is proposed, which incorporates the high-order TMC and the high-order MTRC correction. First, the slant range model for spaceborne ISAR imaging is derived as a third-order polynomial form. Based on this model, a modified high-order TMC method is proposed, which utilizes the minimum entropy method combined with the Broyden – Fletcher – Goldfarb – Shanno (BFGS) algorithm. The proposed method leverages orbital model information to obtain the initial TMC coefficients, enabling TMC to be completed more accurately and effectively. Then, a modified polynomial keystone transform (PKT) is proposed to correct the high-order MTRC. The PKT-based method effectively addresses both the high-order and first-order MTRC corrections simultaneously. Finally, a spaceborne ISAR imaging approach is proposed to obtain well-focused ISAR images. Simulated and real data results validate the analysis of the spaceborne ISAR imaging model and demonstrate the effectiveness of the spaceborne imaging approach for space targets proposed in this paper. [ABSTRACT FROM AUTHOR]

Pine wilt disease (PWD), caused by pine wood nematodes, has brought a great loss in ecology and economy all over the world. In China, the forest health status is also significantly affected by PWD since 1980, especially in coniferous forests and mixed forest regions. The PWD spreads very fast and can cause a healthy pine wood tree to die within a very short time. An effective way to protect other healthy pine wood trees is to discover PWD trees early. Using unmanned aerial vehicle (UAV) images can help people discover the PWD tree quickly and accurately, and a few automatical methods have been developed to monitor PWD trees including the deep learning methods. Because of the robust spatial-temporal transferability, deep learning methods have become the mainstream algorithms to monitor PWD trees. As we know, the training dataset is the most important material to train a deep learning model. However, there is still a lack of a PWD segmentation dataset so far. To fill this gap, in this paper, we have generated the first open-sourced PWD segmentation dataset based on very high-resolution UAV images to help the community conduct PWD monitor research conveniently. This dataset has 994 training samples, and each sample has visible bands with 512 × 512 pixels, and the spatial resolution of this dataset is 0.05 m. In order to train an advanced segmentation model, we have designed a lightweight deep-learning model for mobile devices or edge devices in this manuscript, named MobileSeg. The main feature of MobileSeg is its decoupling, which uses the re-parameterization technology to improve the model performance. Finally, a large-scale real-world scenario experiment with very high-resolution UAV images was utilized to validate the performance of MobileSeg, the experiment result indicated that MobileSeg has achieved the best performance compared with the recent lightweight segmentation models, and the experiment also proved the effectiveness of the proposed PWD dataset. [ABSTRACT FROM AUTHOR]

Multisource remote sensing images have rich features and high interpretability and are widely employed in many applications. However, highly unbalanced category distributions and complex backgrounds have created some difficulties in the application of remote sensing image semantic segmentation tasks, such as low accuracy of small target segmentation and inaccurate edge extraction. To solve these problems, in this paper, a feature map segmentation reconstruction module and an attention upsampling module are proposed. In the encoder part, the input feature map is equally segmented, and the segmented feature map is enlarged to effectively improve the small target feature information expression ability in the model. In the decoder part, the key segmentation and location information of shallow features are obtained using the global view. The deep semantic information and shallow spatial location information are fully combined to achieve a more refined upsampling operation. In addition, the attention mechanism of the spatial and channel squeeze and excitation block (scSE) is applied to pay more attention to important features and to suppress irrelevant background and redundant information. To verify the effectiveness of the proposed method, the WHU-OPT-SAR dataset and six state-of-the-art algorithms are utilized in comparative experiments. The experimental results show that our model has demonstrated the best performance and low computational complexity. With only approximately half the floating-point operation count and the number of model parameters of the MCANet model, which is specially designed for the dataset, our model surpasses MCANet by 1.52% and 1.53% in terms of mean intersection over union (mIoU) and F1 score, respectively. In particular, for small object regions such as roads and other categories, compared to the baseline model, the IoU and F1 score of our model are improved by 5.27% and 3.99% and by 5.68% and 5.65%, respectively. These results demonstrate the superior performance of our model in terms of accuracy and efficiency. [ABSTRACT FROM AUTHOR]