Search

Hyperspectral and multispectral image databases, urban area imagery dataset, point cloud data, aerial video, thermal imaging, and SAR data have been used in experiments. In detail, the authors propose a multi-scaled column-spatial correction network (CSCNet) where the local structural characteristic of the noise and the image global contextual information are jointly exploited at multiple feature scales. The authors have proposed an LST reconstruction method that combines data decomposition with data prediction to obtain spatially and temporally continuous LST data. The peculiarity of the paper is that the authors consider deep neural network-based models, with special attention being focused upon generative adversarial networks, and taking into account the fact that HS images possess abundant spectral information. [Extracted from the article]

To ensure the long-term stable and uninterrupted service of satellite navigation systems, the robustness and reliability of time–frequency systems are crucial. Integrity monitoring is an effective method to enhance the robustness and reliability of time–frequency systems. Time–frequency signals are fundamental for integrity monitoring, with their time differences and frequency biases serving as essential indicators. These indicators are influenced by the inherent characteristics of the time–frequency signals, as well as the links and equipment they traverse. Meanwhile, existing research primarily focuses on only monitoring the integrity of the time–frequency signals' output by the atomic clock group, neglecting the integrity monitoring of the time–frequency signals generated and distributed by the time–frequency signal generation and distribution subsystem. This paper introduces a time–frequency signal integrity monitoring algorithm based on the temperature compensation frequency bias combination model. By analyzing the characteristics of time difference measurements, constructing the temperature compensation frequency bias combination model, and extracting and monitoring noise and frequency bias features from the time difference measurements, the algorithm achieves comprehensive time–frequency signal integrity monitoring. Experimental results demonstrate that the algorithm can effectively detect, identify, and alert users to time–frequency signal faults. Additionally, the model and the integrity monitoring parameters developed in this paper exhibit high adaptability, making them directly applicable to the integrity monitoring of time–frequency signals across various links. Compared with traditional monitoring algorithms, the algorithm proposed in this paper greatly improves the effectiveness, adaptability, and real-time performance of time–frequency signal integrity monitoring. [ABSTRACT FROM AUTHOR]

Pseudo-satellite technology has excellent compatibility with the BDS satellite navigation system in terms of signal systems. It can serve as a stable and reliable positioning signal source in signal-blocking environments. User terminals can achieve continuous high-precision positioning both indoors and outdoors without any modification to the navigation module. As a result, pseudo-satellite indoor positioning has gradually emerged as a research hotspot in the field. However, due to the complex and variable indoor radio propagation environment, signal propagation is interfered with by noise, multipath, non-line-of-sight (NLOS) propagation, etc. The geometric relation-based localization algorithm cannot be applied in indoor non-line-of-sight environments. Therefore, this paper proposes a pseudo-satellite fingerprint localization method based on the discriminative deep belief networks (DDBNs). The method acquires the model parameters of pseudo-satellite multi-carrier noise density signal strength in non-line-of-sight indoor spaces through a greedy unsupervised learning method and gradient descent-supervised learning method. It establishes a mapping relationship between the implied features of the pseudo-satellite multi-carrier noise density signal strength and indoor location, enabling pseudo-satellite fingerprint matching localization in indoor non-line-of-sight environments. In this paper, the performance of the positioning algorithm is verified in dynamic and static scenarios through numerous experiments in a laboratory environment. Compared to the commonly used localization algorithms based on fingerprint library matching, the results demonstrate that, in indoor non-line-of-sight test conditions, the system's 2D static positioning has a maximum error of less than 0.24 m, an RMSE better than 0.12 m, and a 2σ (95.4%) positioning error better than 0.19 m. For 2D dynamic positioning, the maximum error is less than 0.36 m, the average error is 0.23 m, and the 2σ positioning error is better than 0.26 m. These results effectively tackle the challenge of pseudo-satellite indoor positioning in non-line-of-sight environments. [ABSTRACT FROM AUTHOR]

The canopy height model (CHM) derived from LiDAR point cloud data is usually used to accurately identify the position and the canopy dimension of single tree. However, local invalid values (also called data pits) are often encountered during the generation of CHM, which results in low-quality CHM and failure in the detection of treetops. For this reason, this paper proposes an innovative method, called "pixels weighted differential gradient", to filter these data pits accurately and improve the quality of CHM. First, two characteristic parameters, gradient index (GI) and Z-score value (ZV) are extracted from the weighted differential gradient between the pit pixels and their eight neighbors, and then GIs and ZVs are commonly used as criterion for initial identification of data pits. Secondly, CHMs of different resolutions are merged, using the image processing algorithm developed in this paper to distinguish either canopy gaps or data pits. Finally, potential pits were filtered and filled with a reasonable value. The experimental validation and comparative analysis were carried out in a coniferous forest located in Triangle Lake, United States. The experimental results showed that our method could accurately identify potential data pits and retain the canopy structure information in CHM. The root-mean-squared error (RMSE) and mean bias error (MBE) from our method are reduced by between 73% and 26% and 76% and 28%, respectively, when compared with six other methods, including the mean filter, Gaussian filter, median filter, pit-free, spike-free and graph-based progressive morphological filtering (GPMF). The average F1 score from our method could be improved by approximately 4% to 25% when applied in single-tree extraction. [ABSTRACT FROM AUTHOR]

Appropriate environmental sensing methods and visualization representations are crucial foundations for the in situ exploration of planets. In this paper, we developed specialized visualization methods to facilitate the rover's interaction and decision-making processes, as well as to address the path-planning and obstacle-avoidance requirements for lunar polar region exploration and Mars exploration. To achieve this goal, we utilize simulated lunar polar regions and Martian environments. Among them, the lunar rover operating in the permanently shadowed region (PSR) of the simulated crater primarily utilizes light detection and ranging (LiDAR) for environmental sensing; then, we reconstruct a mesh using the Poisson surface reconstruction method. After that, the lunar rover's traveling environment is represented as a red-green-blue (RGB) image, a slope coloration image, and a theoretical water content coloration image, based on different interaction needs and scientific objectives. For the rocky environment where the Mars rover is traveling, this paper enhances the display of the rocks on the Martian surface. It does so by utilizing depth information of the rock instances to highlight their significance for the rover's path-planning and obstacle-avoidance decisions. Such an environmental sensing and enhanced visualization approach facilitates rover path-planning and remote–interactive operations, thereby enabling further exploration activities in the lunar PSR and Mars, in addition to facilitating the study and communication of specific planetary science objectives, and the production and display of basemaps and thematic maps. [ABSTRACT FROM AUTHOR]

Visual object tracking is a key technology that is used in unmanned aerial vehicles (UAVs) to achieve autonomous navigation. In recent years, with the rapid development of deep learning, tracking algorithms based on Siamese neural networks have received widespread attention. However, because of complex and diverse tracking scenarios, as well as limited computational resources, most existing tracking algorithms struggle to ensure real-time stable operation while improving tracking performance. Therefore, studying efficient and fast-tracking frameworks, and enhancing the ability of algorithms to respond to complex scenarios has become crucial. Therefore, this paper proposes a cross-parallel attention and efficient match transformer for aerial tracking (SiamEMT). Firstly, we carefully designed the cross-parallel attention mechanism to encode global feature information and to achieve cross-dimensional interaction and feature correlation aggregation via parallel branches, highlighting feature saliency and reducing global redundancy information, as well as improving the tracking algorithm's ability to distinguish between targets and backgrounds. Meanwhile, we implemented an efficient match transformer to achieve feature matching. This network utilizes parallel, lightweight, multi-head attention mechanisms to pass template information to the search region features, better matching the global similarity between the template and search regions, and improving the algorithm's ability to perceive target location and feature information. Experiments on multiple drone public benchmark tests verified the accuracy and robustness of the proposed tracker in drone tracking scenarios. In addition, on the embedded artificial intelligence (AI) platform AGX Xavier, our algorithm achieved real-time tracking speed, indicating that our algorithm can be effectively applied to UAV tracking scenarios. [ABSTRACT FROM AUTHOR]

In view of the ecological threat posed by cyanobacteria blooms in Taihu Lake (China), this paper presents a study on the area of cyanobacteria extent based on MODIS data using the quantum particle swarm optimization–random forest (QPSO-RF) machine learning algorithm. This paper selects multiple remote sensing input indices that can represent the characteristics of the primary underlying type in Taihu Lake. The proposed method performs best, with an F1 score of 0.91–0.98. Based on this method, the spatio-temporal variation of cyanobacteria blooms in the Taihu Lake complex was analyzed. During 2010–2022, the average area of cyanobacteria blooms in Taihu Lake increased slightly. Severe-scale cyanobacteria blooms occurred in 2015–2019. Cyanobacteria blooms were normally concentrated from May to November. However, the most prolonged extended duration occurred in 2017, lasting for eight months. Spatially, cyanobacteria blooms were mainly identified in the northwestern part of Taihu Lake, with an average occurrence frequency of about 10.0%. The cyanobacteria blooms often began to grow in the northwestern part of the lake and then spread to the Center of the Lake, and also dissipated earliest in the northwestern part of the lake. Our study is also beneficial for monitoring the growth of cyanobacteria blooms in other similar large lakes in long time series. [ABSTRACT FROM AUTHOR]

Ultra-long-distance and high-precision lightning location technology is an important means to realize low-cost and wide-area lightning detection. This paper carried out research on the high-precision location technology of very-low-frequency (VLF) lightning electromagnetic pulse based on the Asia-Pacific Lightning Location Network (APLLN) deployed in 2018. Two key technologies are proposed in this paper: one is the calculation method of signal arrival time using very-low-frequency lightning electromagnetic pulse waveform, and the other is the compression transmission technology of lightning electromagnetic pulse waveform based on a signal principal component analysis. The results of a comparison and evaluation of the improved APLLN with the ADTD system show that the APLLN has a relative location efficiency of 69.1% and an average location error within the network of 4.5 km. [ABSTRACT FROM AUTHOR]

High-resolution and high-squint spaceborne spotlight synthetic aperture radar (SAR) has significant potential for extensive application in remote sensing, but its swath width effectiveness is constrained by a critical factor: severe range cell migration (RCM). To address this, pulse repetition interval (PRI) variation offers a practical scheme for raw data reception. However, the current designs for continuously varying PRI (CV-PRI) exhibit high complexity in engineering. In response to the issue, this paper proposes a novel strategy of stepwise varying PRI (SV-PRI), which demonstrates higher reconstruction accuracy compared with CV-PRI. Furthermore, confronting the azimuth non-uniform sampling characteristics induced by the PRI variation, this paper introduces a complete uniform reconstruction processing based on the azimuth partitioning methodology, which effectively alleviates the inherent contradiction between resolution and swath width. The processing flow, utilizing the temporal point remapping (TPR) concept, ensures the uniformity and coherence of dataset partitioning and reassembly in the context of the interpolation on non-uniform grids. Finally, according to the simulation results, the point target data, processed through the processing flow proposed in this study, have demonstrated effective focusing results. [ABSTRACT FROM AUTHOR]

This document explores the significance of remote sensing in monitoring maritime activities and identifying vessels. It emphasizes the need for surveillance to ensure safety, security, and emergency management, given the increasing number of vessels worldwide. The document highlights the use of technologies like the Automatic Identification System (AIS) and remote sensing in situations where collaborative systems are not reliable. It also discusses the integration of data from different sensors and the application of data science techniques for a comprehensive assessment of maritime traffic. The document concludes by summarizing research papers on ship detection, tracking, and classification using various sensors and data processing techniques. [Extracted from the article]

High-resolution remote sensing image-based vegetation monitoring is a hot topic in remote sensing technology and applications. However, when facing large-scale monitoring across different sensors in broad areas, the current methods suffer from fragmentation and weak generalization capabilities. To address this issue, this paper proposes a multisource high-resolution remote sensing image-based vegetation extraction method that considers the comprehensive perception of multiple features. First, this method utilizes a random forest model to perform feature selection for the vegetation index, selecting an index that enhances the otherness between vegetation and other land features. Based on this, a multifeature synthesis perception convolutional network (MSCIN) is constructed, which enhances the extraction of multiscale feature information, global information interaction, and feature cross-fusion. The MSCIN network simultaneously constructs dual-branch parallel networks for spectral features and vegetation index features, strengthening multiscale feature extraction while reducing the loss of detailed features by simplifying the dense connection module. Furthermore, to facilitate global information interaction between the original spectral information and vegetation index features, a dual-path multihead cross-attention fusion module is designed. This module enhances the differentiation of vegetation from other land features and improves the network's generalization performance, enabling vegetation extraction from multisource high-resolution remote sensing data. To validate the effectiveness of this method, we randomly selected six test areas within Anhui Province and compared the results with three different data sources and other typical methods (NDVI, RFC, OCBDL, and HRNet). The results demonstrate that the MSCIN method proposed in this paper, under the premise of using only GF2 satellite images as samples, exhibits robust accuracy in extraction results across different sensors. It overcomes the rapid degradation of accuracy observed in other methods with various sensors and addresses issues such as internal fragmentation, false positives, and false negatives caused by sample generalization and image diversity. [ABSTRACT FROM AUTHOR]

Mobile LiDAR technology is a powerful tool that accurately captures spatial information about typical static objects in road scenes. However, the precise extraction and classification of these objects pose persistent technical challenges. In this paper, we employ a deep learning approach to tackle the point cloud classification problem. Despite the popularity of the PointNet++ network for direct point cloud processing, it encounters issues related to insufficient feature learning and low accuracy. To address these limitations, we introduce a novel layer-wise optimization network, LO-Net. Initially, LO-Net utilizes the set abstraction module from PointNet++ to extract initial local features. It further enhances these features through the edge convolution capabilities of GraphConv and optimizes them using the "Unite_module" for semantic enhancement. Finally, it employs a point cloud spatial pyramid joint pooling module, developed by the authors, for the multiscale pooling of final low-level local features. Combining three layers of local features, LO-Net sends them to the fully connected layer for accurate point cloud classification. Considering real-world scenarios, road scene data often consist of incomplete point cloud data due to factors such as occlusion. In contrast, models in public datasets are typically more complete but may not accurately reflect real-world conditions. To bridge this gap, we transformed road point cloud data collected by mobile LiDAR into a dataset suitable for network training. This dataset encompasses nine common road scene features; hence, we named it the Road9 dataset and conducted classification research based on this dataset. The experimental analysis demonstrates that the proposed algorithm model yielded favorable results on the public datasets ModelNet40, ModelNet10, and the Sydney Urban Objects Dataset, achieving accuracies of 91.2%, 94.2%, and 79.5%, respectively. On the custom road scene dataset, Road9, the algorithm model proposed in this paper demonstrated outstanding classification performance, achieving a classification accuracy of 98.5%. [ABSTRACT FROM AUTHOR]

Ship-radiated noise is the main basis for ship detection in underwater acoustic environments. Due to the increasing human activity in the ocean, the captured ship noise is usually mixed with or covered by other signals or noise. On the other hand, due to the softening effect of bubbles in the water generated by ships, ship noise undergoes non-negligible nonlinear distortion. To mitigate the nonlinear distortion and separate the target ship noise, blind source separation (BSS) becomes a promising solution. However, underwater acoustic nonlinear models are seldom used in research for nonlinear BSS. This paper is based on the hypothesis that the recovery and separation accuracy can be improved by considering this nonlinear effect in the underwater environment. The purpose of this research is to explore and discover a method with the above advantages. In this paper, a model is used in underwater BSS to describe the nonlinear impact of the softening effect of bubbles on ship noise. To separate the target ship-radiated noise from the nonlinear mixtures, an end-to-end network combining an attention mechanism and bidirectional long short-term memory (Bi-LSTM) recurrent neural network is proposed. Ship noise from the database ShipsEar and line spectrum signals are used in the simulation. The simulation results show that, compared with several recent neural networks used for linear and nonlinear BSS, the proposed scheme has an advantage in terms of the mean square error, correlation coefficient and signal-to-distortion ratio. [ABSTRACT FROM AUTHOR]

When Synthetic Aperture Radar (SAR) observes complex structural targets such as oil tanks, it is easily interfered with by multipath signals, resulting in a large number of multipath ghost images in the SAR image, which seriously affect the image clarity. To address this problem, this paper proposes a multi-aspect multipath suppression method. This method observes complex structural targets from different azimuth angles to obtain a multi-aspect image sequence and then uses the difference in sequence features between the target image and the multipath ghost image with respect to aspect angle to separate them. This paper takes a floating-roof oil tank as an example to analyze the propagation path and the ghost image characteristics of multipath signals under different observation aspects. We conclude that the scattering center of the multipath ghost image changes with the radar observation aspect, whereas the scattering center of the target image does not. This paper uses the Robust Principal Component Analysis (RPCA) method to decompose the image sequence matrix into two parts: a sparse matrix and a low-rank matrix. The low-rank matrix represents the aspect-stable principal component in the image sequence; that is, the real scattering center. The sparse matrix represents the part of the image sequence that deviates from the principal component; that is, the signal that varies with aspect, mainly including multipath signals, sidelobes, anisotropic signals, etc. By reconstructing the low-rank matrix and the sparse matrix, respectively, we can obtain the image after multipath signal suppression and also the multipath ghost image. Both the target and the multipath signal provide useful information. The image after multipath signal suppression is useful for obtaining the structural information of the target, and the multipath ghost image is useful for analyzing the multipath phenomenon of the complex structure target. This paper conducts experimental verification using real airborne SAR data of an external floating roof oil tank and compares three methods: RPCA, PCA, and sub-aperture fusion method. The experiment shows that the RPCA method can better separate the target image and the multipath ghost image. [ABSTRACT FROM AUTHOR]

A skeletal pose estimation method, named RVRU-Pose, is proposed to estimate the skeletal pose of vulnerable road users based on distributed non-coherent mmWave radar. In view of the limitation that existing methods for skeletal pose estimation are only applicable to small scenes, this paper proposes a strategy that combines radar intensity heatmaps and coordinate heatmaps as input to a deep learning network. In addition, we design a multi-resolution data augmentation and training method suitable for radar to achieve target pose estimation for remote and multi-target application scenarios. Experimental results show that RVRU-Pose can achieve better than 2 cm average localization accuracy for different subjects in different scenarios, which is superior in terms of accuracy and time compared to existing state-of-the-art methods for human skeletal pose estimation with radar. As an essential performance parameter of radar, the impact of angular resolution on the estimation accuracy of a skeletal pose is quantitatively analyzed and evaluated in this paper. Finally, RVRU-Pose has also been extended to the task of estimating the skeletal pose of a cyclist, reflecting the strong scalability of the proposed method. [ABSTRACT FROM AUTHOR]

Circular synthetic aperture radar (CSAR) possesses the capability of multi-angle observation, breaking through the geometric observation constraints of traditional strip SAR and holding the potential for three-dimensional imaging. Its sub-wavelength level of planar resolution, resulting from a long synthetic aperture, makes CSAR highly valuable in the field of high-precision mapping. However, the motion geometry of CSAR is more intricate compared to traditional strip SAR, demanding high precision from navigation systems. The accumulation of errors over the long synthetic aperture time cannot be overlooked. CSAR exhibits significant coupling between the range and azimuth directions, making traditional motion compensation methods based on linear SAR unsuitable for direct application in CSAR. The dynamic nature of flight, with its continuous changes in attitude, introduces a significant deformation error between the non-rigidly connected Inertial Measurement Unit (IMU) and the Global Positioning System (GPS). This deformation error makes it difficult to accurately obtain radar position information, resulting in imaging defocus. The research in this article uncovers a correlation between the deformation error and radial acceleration. Leveraging this insight, we propose utilizing radial acceleration to estimate residual motion errors. This paper delves into the analysis of Position and Orientation System (POS) errors, presenting a novel high-resolution CSAR motion compensation method based on airborne platform acceleration information. Once the system deformation parameters are calibrated using point targets, the deformation error can be directly calculated and compensated based on the acceleration information, ultimately resulting in the generation of a high-resolution image. In this paper, the effectiveness of the method is verified with airborne flight test data. This method can compensate for the deformation error and effectively improve the peak sidelobe ratio and integral sidelobe ratio of the target, thus improving image quality. The introduction of acceleration information provides new means and methods for high-resolution CSAR imaging. [ABSTRACT FROM AUTHOR]

In this paper, we propose a method for extracting the structure of an indoor environment using radar. When using the radar in an indoor environment, ghost targets are observed through the multipath propagation of radio waves. The presence of these ghost targets obstructs accurate mapping in the indoor environment, consequently hindering the extraction of the indoor environment. Therefore, we propose a deep learning-based method that uses image-to-image translation to extract the structure of the indoor environment by removing ghost targets from the indoor environment map. In this paper, the proposed method employs a conditional generative adversarial network (CGAN), which includes a U-Net-based generator and a patch-generative adversarial network-based discriminator. By repeating the process of determining whether the structure of the generated indoor environment is real or fake, CGAN ultimately returns a structure similar to the real environment. First, we generate a map of the indoor environment using radar, which includes ghost targets. Next, the structure of the indoor environment is extracted from the map using the proposed method. Then, we compare the proposed method, which is based on the structural similarity index and structural content, with the k-nearest neighbors algorithm, Hough transform, and density-based spatial clustering of applications with noise-based environment extraction method. When comparing the methods, our proposed method offers the advantage of extracting a more accurate environment without requiring parameter adjustments, even when the environment is changed. [ABSTRACT FROM AUTHOR]

This paper investigates a real-time process generator of wideband signals, which calculates waveforms in a field-programmable gate array (FPGA) using the high-level synthesis (HLS) method. To obtain high-resolution and wide-swath images, the generator must produce multiple modes of large time-bandwidth product (TBP) linear frequency modulation (LFM) signals. However, the conventional storage method is unrealistic as it requires huge storage resources to save pre-computed waveforms. Therefore, this paper proposes a novel processing approach that calculates waveforms in real-time based simply on parameters such as the sampling frequency, bandwidth, and time width. Additionally, this paper implements predistortion through the polynomial curve to approximate phase errors of the system. The parallelizing process in the FPGA is necessary to satisfy the high-speed requirement of a digital-to-analog converter (DAC); however, repeatedly multiplexing real-time calculation consumes extensive logic and DSP resources, potentially exceeding FPGA limitations. To address this, this paper proposes a piecewise linear algorithm to conserve resources, which processes the polynomial only once, acquires the difference in two adjacent values through the register and pipeline, and then adds this increment to facilitate parallel computations. The performance of this proposed generator is validated through simulation and implemented in experiments with an X-band airborne SAR system. [ABSTRACT FROM AUTHOR]

Tree species classification based on multi-source remote sensing data is essential for ecological evaluation, environmental monitoring, and forest management. The optimization of classification features and the performance of classification methods are crucial to tree species classification. This paper proposes Angle-weighted Standard Deviation Elliptic Cross-merge Rate (ASDER) as a separability metric for feature optimization. ASDER uses mutual information to represent the separability metric and avoids the difficulty of differentiation caused by multiple ellipse centers and coordinate origins forming straight lines by angle weighting. In classification method, Multi-head Self-attention Long Short-Term Memory—Full Convolution Network (MALSTM-FCN) is constructed in this paper. MALSTM-FCN enhances the global correlation in time series and improves classification accuracy through a multi-head self-attention mechanism. This paper takes Beijing Olympic Forest Park (after this, referred to as Aosen) as the research area, constructs a tree species classification dataset based on an actual ground survey, and obtains a classification accuracy of 95.20% using the above method. This paper demonstrates the effectiveness of ASDER and MALSTM-FCN by comparing temporal entropy and LSTM-FCN and shows that the method has some practicality for tree species classification. [ABSTRACT FROM AUTHOR]

Object detection is one of the fundamental tasks in computer vision, holding immense significance in the realm of intelligent mobile scenes. This paper proposes a hybrid cross-feature interaction (HCFI) attention module for object detection in intelligent mobile scenes. Firstly, the paper introduces multiple kernel (MK) spatial pyramid pooling (SPP) based on SPP and improves the channel attention using its structure. This results in a hybrid cross-channel interaction (HCCI) attention module with better cross-channel interaction performance. Additionally, we bolster spatial attention by incorporating dilated convolutions, leading to the creation of the cross-spatial interaction (CSI) attention module with superior cross-spatial interaction performance. By seamlessly combining the above two modules, we achieve an improved HCFI attention module without resorting to computationally expensive operations. Through a series of experiments involving various detectors and datasets, our proposed method consistently demonstrates superior performance. This results in a performance improvement of 1.53% for YOLOX on COCO and a performance boost of 2.05% for YOLOv5 on BDD100K. Furthermore, we propose a solution that combines HCCI and HCFI to address the challenge of extremely small output feature layers in detectors, such as SSD. The experimental results indicate that the proposed method significantly improves the attention capability of object detection in intelligent mobile scenes. [ABSTRACT FROM AUTHOR]

Contactless and non-destructive measuring tools can facilitate the moisture monitoring of bagged or bulk grain during transportation and storage. However, accurate target recognition and size prediction always impede the effectiveness of contactless monitoring in actual use. This paper developed a novel 3D reconstruction method upon multi-angle point clouds using a binocular depth camera and a proper Yolo-based neural model to resolve the problem. With this method, this paper developed an embedded and low-cost monitoring system for the in-warehouse grain bags, which predicted targets' 3D size and boosted contactless grain moisture measuring. Identifying and extracting the object of interest from the complex background was challenging in size prediction of the grain silo-bag on a conveyor. This study first evaluated a series of Yolo-based neural network models and explored the most appropriate neural network structure for accurately extracting the grain bag. In point-cloud processing, this study constructed a rotation matrix to fuse multi-angle point clouds to generate a complete one. This study deployed all the above methods on a Raspberry Pi-embedded board to perform the grain bag's 3D reconstruction and size prediction. For experimental validation, this study built the 3D reconstruction platform and tested grain bags' reconstruction performance. First, this study determined the appropriate positions (−60°, 0°, 60°) with the least positions and high reconstruction quality. Then, this study validated the efficacy of the embedded system by evaluating its speed and accuracy and comparing it to the original Torch model. Results demonstrated that the NCNN-accelerated model significantly enhanced the average processing speed, nearly 30 times faster than the Torch model. The proposed system predicted the objects' length, width, and height, achieving accuracies of 97.76%, 97.02%, and 96.81%, respectively. The maximum residual value was less than 9 mm. And all the root mean square errors were less than 7 mm. In the future, the system will mount three depth cameras for achieving real-time size prediction and introduce a contactless measuring tool to finalize grain moisture detection. [ABSTRACT FROM AUTHOR]

In this paper, we investigate deep-learning-based image inpainting techniques for emergency remote sensing mapping. Image inpainting can generate fabricated targets to conceal real-world private structures and ensure informational privacy. However, casual inpainting outputs may seem incongruous within original contexts. In addition, the residuals of original targets may persist in the hiding results. A Residual Attention Target-Hiding (RATH) model has been proposed to address these limitations for remote sensing target hiding. The RATH model introduces the residual attention mechanism to replace gated convolutions, thereby reducing parameters, mitigating gradient issues, and learning the distribution of targets present in the original images. Furthermore, this paper modifies the fusion module in the contextual attention layer to enlarge the fusion patch size. We extend the edge-guided function to preserve the original target information and confound viewers. Ablation studies on an open dataset proved the efficiency of RATH for image inpainting and target hiding. RATH had the highest similarity, with a 90.44 % structural similarity index metric (SSIM), for edge-guided target hiding. The training parameters had 1 M fewer values than gated convolution (Gated Conv). Finally, we present two automated target-hiding techniques that integrate semantic segmentation with direct target hiding or edge-guided synthesis for remote sensing mapping applications. [ABSTRACT FROM AUTHOR]

With the rapid development of multi-mode and multi-frequency GNSSs (including GPS, GLONASS, BDS, Galileo, and QZSS), more observations for research on ionosphere can be provided. The Global Ionospheric Map (GIM) products are generated based on the observation of multi-mode and multi-frequency GNSSs, and comparisons with other GIMs provided by the ionosphere analysis centers are provided in this paper. Taking the CODE (Center of Orbit Determination in Europe) GIM as a reference during 30 days in January 2019, for the GIMs from JPL (Jet Puls Laboratory), UPC (Technical University of Catalonia), ESA (European Space Agency), WHU (Wuhan University), CAS (Chinese Academy of Sciences), and MMG (The multi-mode and multi-frequency GNSS observations used in this paper), the mean bias with respect to CODE products is 1.87, 1.30, −0.10, 0.01, −0.02, and −0.71 TECu, and the RMS is 2.12, 2.00, 1.33, 0.88, 0.88, and 1.30 TECu, respectively. The estimated multi-type DCB is also in good agreement with the DCB products provided by the MGEX. [ABSTRACT FROM AUTHOR]

Deep learning has gained widespread interest in the task of building semantic segmentation modelling using remote sensing images; however, neural network models require a large number of training samples to achieve better classification performance, and the models are more sensitive to error patches in the training samples. The training samples obtained in semi-supervised classification methods need less reliable weakly labelled samples, but current semi-supervised classification research puts the generated weak samples directly into the model for applications, with less consideration of the impact of the accuracy and quality improvement of the weak samples on the subsequent model classification. Therefore, to address the problem of generating and optimising the quality of weak samples from training data in deep learning, this paper proposes a semi-supervised building classification framework. Firstly, based on the test results of the remote sensing image segmentation model and the unsupervised classification results of LiDAR point cloud data, this paper quickly generates weak image samples of buildings. Secondly, in order to improve the quality of the spots of the weak samples, an iterative optimisation strategy of the weak samples is proposed to compare and analyse the weak samples with the real samples and extract the accurate samples from the weak samples. Finally, the real samples, the weak samples, and the optimised weak samples are input into the semantic segmentation model of buildings for accuracy evaluation and analysis. The effectiveness of this paper's approach was experimentally verified on two different building datasets, and the optimised weak samples improved by 1.9% and 0.6%, respectively, in the test accuracy mIoU compared to the initial weak samples. The results demonstrate that the semi-supervised classification framework proposed in this paper can be used to alleviate the model's demand for a large number of real-labelled samples while improving the ability to utilise weak samples, and it can be used as an alternative to fully supervised classification methods in deep learning model applications that require a large number of training samples. [ABSTRACT FROM AUTHOR]

Currently, there is a wealth of data and expert knowledge available on monitoring agro-meteorological disasters. However, there is still a lack of technical means to organically integrate and analyze heterogeneous data sources in a collaborative manner. This paper proposes a method for monitoring agro-meteorological disasters based on a spatio-temporal knowledge graph. It employs a semantic ontology framework to achieve the organic fusion of multi-source heterogeneous data, including remote sensing data, meteorological data, farmland data, crop information, etc. And it formalizes expert knowledge and computational models into knowledge inference rules, thereby enabling monitoring, early warning, and disaster analysis of agricultural crops within the observed area. The experimental area for this research is the wheat planting region in three counties in Henan Province. The method is tested using simulation monitoring, early warning, and impact calculation of the past two occurrences of dry hot wind disasters. The experimental results demonstrate that the proposed method can provide more specific and accurate warning information and post-disaster analysis results compared to raw records. The statistical results of NDVI decline also validate the correlation between the severity of wheat damage caused by dry hot winds and the intensity and duration of their occurrences. Regarding remote sensing data, this paper proposes a method that directly incorporates remote sensing data into spatio-temporal knowledge inference calculations. By integrating remote sensing data into the regular monitoring process, the advantages of remote sensing data granted by continuous observation are utilized. This approach represents a beneficial attempt to organically integrate remote sensing and meteorological data for monitoring, early warning, and evaluation analysis of agro-meteorological disasters. [ABSTRACT FROM AUTHOR]

In order to obtain high-quality images, the application of low-illumination image enhancement techniques plays a vital role in enhancing the overall visual appeal. However, it is particularly difficult to enhance an image while maintaining the original information of the scene. The augmentation method based on Retinex theory is widely considered as one of the representative techniques for such problems, but this method still has some limitations. First of all, noise is easily ignored in the process of model building, and the robustness of the model needs to be improved. Secondly, the image decomposition is less effective, so that part of the image information is not effectively presented. Finally, the optimization procedure is computationally complicated. This paper introduces a novel approach for enhancing low-illumination images by utilizing the relative deviation of local gradients. The proposed method aims to address the challenges associated with low-illumination images and offers a solution to these issues. In this paper, local gradient relative deviation is used as a constraint term and a noise term is added to highlight the image texture and structure and improve the robustness of the models, considering that L P achieves piecewise smoothing with better sparsity compared to the sum norm commonly used by L 1 and L 2 norms. In this paper, the L 2 − L P norm is used to constrain the model, which smooths the illumination component and better preserves the details of the reflectance component. In addition, to efficiently solve the optimization problem, the alternating direction multiplier method is chosen to transform the optimization process into the solution of several sub-problems. In comparison to traditional Retinex models, the proposed method excels in its ability to simultaneously enhance the image and suppress noise effectively. The experimental outcomes demonstrate the effectiveness of the proposed model in enhancing both simulated and real data. This approach can be applied to low-illumination remote sensing images to obtain high-quality remote sensing image data. [ABSTRACT FROM AUTHOR]

At present, a variety of global remote sensing surface downwelling longwave radiation (SDL) products are used for atmospheric science research; however, there are few studies on the quantitative evaluation of differences among different SDL products. In order to evaluate the differences among different SDL products quantitatively, we have selected three commonly used SDL products—Clouds and the Earth's Radiant Energy System-Synoptic Radiative Fluxes and Clouds (CERES-SYN), the European Centre for Medium Range Weather Forecasts-Surface Radiation Budget (ECMWF-SRB) and the Global Energy and Water Exchanges Project-Surface Radiation Budget (GEWEX-SRB)—to comprehensively study in this paper. The results show that there are significant differences among the three SDL products in some areas, such as in the Arctic, the Antarctic, the Sahara, the Tibet Plateau, and Greenland. The maximum absolute root mean square error (RMSEab) in these areas is greater than 20 Wm−2, the maximum relative root mean square error (RMSEre) is greater than 20%, the maximum and minimum absolute mean bias error (MBEab) are about 20 Wm−2 and −20 Wm−2, respectively, and the maximum and minimum relative mean bias error (MBEre) are about 10% and −10%, respectively. Among the three SDL products, the difference between the ECMWF-SRB and GEWEX-SRB is the most significant. In addition, this paper also analyzed the differences among different SDL products based on three aspects. Firstly, the differences among the three SDL products show that there is significant seasonality, and the differences among different months may vary greatly. However, the differences are not sensitive to years. Secondly, there are some differences in cloud-forcing radiative fluxes (CFRFs) of different SDL products, which is also an important factor affecting the difference between different SDL products. Finally, in the process of converting high temporal resolution SDL products into monthly SDL products, data processing also affects the difference between different SDL products. [ABSTRACT FROM AUTHOR]

This paper studies a coalition game theoretic power allocation algorithm for multi-target detection in radar networks based on low probability of intercept (LPI). The main goal of the algorithm is to reduce the total radiated power of the radar networks while satisfying the predetermined target detection performance of each radar. Firstly, a utility function that comprehensively considers both target detection performance and the radiated power of the radar networks is designed with LPI performance as the guiding principle. Secondly, it causes a coalition to form between cooperating radars, and radars within the same coalition share information. On this basis, a mathematical model for power allocation in radar networks based on coalition game theory is established. The model takes the given target detection performance as a constraint and maximizing system energy efficiency and optimal power allocation as the optimization objective. Furthermore, this paper proposes a game algorithm for joint coalition formation and power allocation in a multi-target detection scenario. Finally, the existence and uniqueness of the Nash equilibrium (NE) solution are proven through strict mathematical deduction. Simulation results validate the effectiveness and feasibility of the proposed algorithm. [ABSTRACT FROM AUTHOR]

This research is part of a wider framework of index literature studies that have been conducted in the past few years. Some of these have had a focus on specific remote sensing (RS) technologies, while others have tackled specific threats to cultural heritage and landscapes. By considering both damages to heritage sites and technologies used for documentation and the monitoring of such occurrences, this paper unveils the current trends on a global scale in the study of the threats to heritage caused by both human-induced and natural hazards. Papers published by Europe-based researchers over the last 20 years using RS and Earth Observation (EO) techniques were surveyed alongside recommendations and programmatic documents issued by institutions in charge of heritage protection and management of several countries in Europe. Around 300 documents, including scientific articles (published from 2000 until 2022) and Grey literature (from 2008 and 2022), were analysed. The data collection and analysis were undertaken by a working group that was intentionally composed to bring together diverse perspectives and expertise, i.e., requirements of heritage professionals using RS and EO technologies, knowledge on technologies and their use in the field, and expertise in methodology implementation to support heritage management. The results highlight the type of hazards considered the most and the geographical distribution of the archaeological sites and monuments targeted by these studies; the countries the researchers are affiliated with; the types of RS and specifically satellite-based technologies used (and hence the type of data used); the tendencies of satellite data usage—visual interpretation, image processing, employment of machine learning, and AI; the technologies most applied by public institutions and practitioners; and many others. Recommendations and future trajectories are then outlined to efficiently reframe discrepancies between types of damage that have received the greatest attention in the literature and the most impactful ones in terms of the number of sites damaged. [ABSTRACT FROM AUTHOR]

The article announce the winners of the periodical's 10th Anniversary Best Paper Award given to "Unmanned Aircraft Systems in Remote Sensing and Scientific Research: Classification and Considerations of Use" in the special award category.

Permafrost is a significant part of the cryosphere, which has gained increasing attention from scientists, policy-makers, and the general public due to global warming, environmental degradation, water shortages, and intense human activities. Although many permafrost research review articles have been published, these studies were predominantly limited to either one subject or one field, while systematic studies about permafrost based on bibliometric analysis methods remain limited. We aim to fill this gap by conducting a bibliometric analysis of 13,697 articles in the field of permafrost research from 1942 to 2021, collected from the Web of Science core collection database. The results indicate that permafrost research is a typically multi-author, multi-country, and multi-institution cooperative field, involved in many research fields. The cumulative number of publications has presented an exponential increase over the past 80 years, with an average annual growth rate of 10.40%. Since 2000, China has seen a rapid growth in the number of publications per year, surpassing the USA in 2016 and leading in the years since then. In addition, the authors from China have great contributions in publications, and there is good room for permafrost development in the future according to the authors' M-index ranking. After the analysis of authors' keywords, we found that, compared to the conventional methods, machine learning and interferometric synthetic aperture radar (InSAR) are new technological approaches introduced in recent years, and the Qinghai–Tibet Plateau has become a popular study area. The results presented here can help related researchers, scholars, and students in the field to better understand the past developments, current status, and future trends of permafrost research. Furthermore, this paper presents and expands the general process of the bibliometric method used in permafrost studies, which can provide researchers with new inspirations and improve discipline research approach. [ABSTRACT FROM AUTHOR]

With the rapid development of the energy industry and the growth of the global energy demand in recent years, the development of the photovoltaic industry has become increasingly significant. However, the development of the PV industry is constrained by high land costs, and land in central cities and industrial areas is often very expensive and unsuitable for the installation of PV equipment in large areas. With this background knowledge, the key to evaluating the PV potential is by counting the rooftop information of buildings, and an ideal solution for extracting building rooftop information is from remote sensing satellite images using the deep learning method; however, the deep learning method often requires large-scale labeled samples, and the labeling of remote sensing images is often time-consuming and expensive. To reduce the burden of data labeling, models trained on large datasets can be used as pre-trained models (e.g., ImageNet) to provide prior knowledge for training. However, most of the existing pre-trained model parameters are not suitable for direct transfer to remote sensing tasks. In this paper, we design a pseudo-label-guided self-supervised learning (PGSSL) semantic segmentation network structure based on high-resolution remote sensing images to extract building information. The pseudo-label-guided learning method allows the feature results extracted by the pretext task to be more applicable to the target task and ultimately improves segmentation accuracy. Our proposed method achieves better results than current contrastive learning methods in most experiments and uses only about 20–50% of the labeled data to achieve comparable performance with random initialization. In addition, a more accurate statistical method for building density distribution is designed based on the semantic segmentation results. This method addresses the last step of the extraction results oriented to the PV potential assessment, and this paper is validated in Beijing, China, to demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

The extensive deployment of wind farms has significantly impacted the detection capabilities of space–air bistatic radar (SABR) systems. Although space–time adaptive processing techniques are available, their performance is significantly degraded, and even unable to suppress clutter. This paper explores the geometric configuration of the SABR system and the selection of detection areas, establishing a space–time clutter model that addresses the effects of wind turbine clutter (WTC). Expressions for spatial and Doppler frequencies have been derived to deeply analyze the characteristics of clutter spreading. Building on this, the paper extends two-dimensional space–time data to three-dimensional azimuth–elevation–Doppler data. It proposes a three-dimensional space–time multi-beam (STMB) strategy incorporating the Ordering Points to Identify the Clustering Structure (OPTICS) clustering algorithm to suppress WTC effectively. This algorithm selects WTC samples and applies OPTICS clustering to the clutter-suppressed data to achieve this effect. Simulation experiments further verify the effectiveness of the algorithm. [ABSTRACT FROM AUTHOR]

Accurate cloud detection is a crucial initial stage in optical satellite remote sensing. In this study, a daytime cloud mask model is proposed for the Advanced Geostationary Radiation Imager (AGRI) onboard the Fengyun 4A (FY-4A) satellite based on a deep learning approach. The model, named "Convolutional and Attention-based Cloud Mask Net (CACM-Net)", was trained using the 2021 dataset with CALIPSO data as the truth value. Two CACM-Net models were trained based on a satellite zenith angle (SZA) < 70° and >70°, respectively. The study evaluated the National Satellite Meteorological Center (NSMC) cloud mask product and compared it with the method established in this paper. The results indicate that CACM-Net outperforms the NSMC cloud mask product overall. Specifically, in the SZA < 70° subset, CACM-Net enhances accuracy, precision, and F1 score by 4.8%, 7.3%, and 3.6%, respectively, while reducing the false alarm rate (FAR) by approximately 7.3%. In the SZA > 70° section, improvements of 12.2%, 19.5%, and 8% in accuracy, precision, and F1 score, respectively, were observed, with a 19.5% reduction in FAR compared to NSMC. An independent validation dataset for January–June 2023 further validates the performance of CACM-Net. The results show improvements of 3.5%, 2.2%, and 2.8% in accuracy, precision, and F1 scores for SZA < 70° and 7.8%, 11.3%, and 4.8% for SZA > 70°, respectively, along with reductions in FAR. Cross-comparison with other satellite cloud mask products reveals high levels of agreement, with 88.6% and 86.3% matching results with the MODIS and Himawari-9 products, respectively. These results confirm the reliability of the CACM-Net cloud mask model, which can produce stable and high-quality FY-4A AGRI cloud mask results. [ABSTRACT FROM AUTHOR]

This paper focuses on detecting and tracking maneuvering weak targets in the range–Doppler (RD) plane with the track-before-detect (TBD) algorithm based on dynamic programming (DP). Traditional DP-TBD algorithms integrate target detection and tracking in their framework while searching the paths provided by a predefined model of the kinematic properties within the constraints allowed. However, both the approximate motion model used in the RD plane and the model mismatch caused when the target undergoes a maneuver can degrade the TBD performance sharply. To address these issues, this paper accurately describes the evolution of the RD equation based on Constant Acceleration (CA) and Coordinated Turn (CT) motion models with the process noise in the Cartesian coordinate system, and it also employs a recursive method to estimate the parameters in the equations for efficient energy accumulation and path searches. Facing the situation that targets energy accumulation during the DP iteration process will lead to an expansion of the target energy accumulation process. This paper designs a more efficient Optimization Function (OF) to inhibit the expansion effect, improve the resolution of the nearby targets, and increase the detection probability of the weak targets simultaneously. In addition, to search the trajectory more efficiently and accurately, we improved the process of DP multi-frame accumulation, thus reducing the computation amount of large-scale searches. Finally, the effectiveness of the proposed method for CA and CT motion target detection and tracking is verified by many of the simulation experiments that were conducted in this paper. [ABSTRACT FROM AUTHOR]

This paper presents exploratory work on the use of Testing with Concept Activation Vectors (TCAV) within a concept-based explanation framework to provide the explainability of millimeter-wave (MMW) radar target recognition. Given that the radar spectrum is difficult for non-domain experts to understand visually, defining concepts for radar remains a significant challenge. In response, drawing from the visual analytical experience of experts, some basic concepts based on brightness, striping, size, and shape are adopted in this paper. However, the simplicity of basic concept definitions sometimes leads to vague correlations with recognition targets and significant variability among individuals, limiting their adaptability to specific tasks. To address these issues, this study proposes a Basic Concept-Guided Deep Embedding Clustering (BCG-DEC) method that can effectively discover task-specific composite concepts. BCG-DEC methodically analyzes the deep semantic information of radar data through four distinct stages from the perspective of concept discovery, ensuring that the concepts discovered accurately conform to the task-specific property of MMW radar target recognition. The experimental results show that the proposed method not only expands the number of concepts but also effectively solves the problem of difficulty in annotating basic concepts. In the ROD2021 MMW radar explainability experiments, the concepts proved crucial for recognizing specific categories of radar targets. [ABSTRACT FROM AUTHOR]

Dust storms play a crucial role in the climate system and the space environment of Mars, significantly impacting human exploration activities on the planet. The Martian dust storms exhibit significant regional, seasonal and interannual variations due to various controlling factors such as large-scale atmospheric circulation, varying solar radiation forcing, and Martial orbital and rotational motions and their coupling to the atmospheric dynamics. This paper aims to review current understandings of Martian dust storms. This paper begins by elucidating the basic properties of dust storms, their driving mechanisms, and their impacts on atmospheric dynamics, atmospheric electric property, space environment, topography, and Mars explorations. The paper then introduces the observation methods on different platforms, including orbiters and landers/rovers, along with datasets constructed based on these historical observations of Martian dust storms. Finally, we propose dust storm monitoring and predicting for the upcoming Chinese Tianwen-3 Mars sample return mission. It concludes by depicting the future research topics aimed at systematically understanding Martian dust storms. [ABSTRACT FROM AUTHOR]

Numerous policies have been proposed by international and supranational institutions, such as the European Union, to surveil Earth from space and furnish indicators of environmental conditions across diverse scenarios. In tandem with these policies, different initiatives, particularly on both sides of the Atlantic, have emerged to provide valuable data for environmental management such as the concept of essential climate variables. However, a key question arises: do the available data align with the monitoring requirements outlined in these policies? In this paper, we concentrate on Earth Observation (EO) optical data applications for environmental monitoring, with a specific emphasis on ocean colour. In a rapidly changing climate, it becomes imperative to consider data requirements for upcoming space missions. We place particular significance on the application of these data when monitoring lakes and marine protected areas (MPAs). These two use cases, albeit very different in nature, underscore the necessity for higher-spatial-resolution imagery to effectively study these vital habitats. Limnological ecosystems, sensitive to ice melting and temperature fluctuations, serve as crucial indicators of a climate in change. Simultaneously, MPAs, although generally small in size, play a crucial role in safeguarding marine biodiversity and supporting sustainable marine resource management. They are increasingly acknowledged as a critical component of global efforts to conserve and manage marine ecosystems, as exemplified by Target 3 of the Kunming–Montreal Global Biodiversity Framework (GBF), which aims to effectively conserve 30% of terrestrial, inland water, coastal, and marine areas by 2030 through protected areas and other conservation measures. In this paper, we analysed different policies concerning EO data and their application to environmental-based monitoring. We also reviewed and analysed the existing relevant literature in order to find gaps that need to be bridged to effectively monitor these habitats in an ecosystem-based approach, making data more accessible, leading to the generation of water quality indicators derived from new high- and very high-resolution satellite monitoring focusing especially on Chlorophyll-a concentrations. Such data are pivotal for comprehending, at small and local scales, how these habitats are responding to climate change and various stressors. [ABSTRACT FROM AUTHOR]

Generative adversarial networks (or GANs) are a specific deep learning architecture often used for different usages, such as data generation or image-to-image translation. In recent years, this structure has gained increased popularity and has been used in different fields. One area of expertise currently in vogue is the use of GANs to produce synthetic aperture radar (SAR) data, and especially expand training datasets for SAR automatic target recognition (ATR). In effect, the complex SAR image formation makes these kind of data rich in information, leading to the use of deep networks in deep learning-based methods. Yet, deep networks also require sufficient data for training. However, contrary to optical images, we generally do not have a substantial number of available SAR images because of their acquisition and labelling cost; GANs are then an interesting tool. Concurrently, how to improve explainability for SAR ATR deep neural networks and how to make their reasoning more transparent have been increasingly explored as model opacity deteriorates trust of users. This paper aims at reviewing how GANs are used with SAR images, but also giving perspectives on how GANs could be used to improve interpretability and explainability of SAR classifiers. [ABSTRACT FROM AUTHOR]

The ionospheric effects in repeat-pass SAR interferometry (InSAR) have become a rising concern with the increasing interest in low-frequency SAR. The ionosphere will introduce serious phase errors in the interferogram, which should be properly corrected. In this paper, the influence of the wavenumber shift on the Range Split-Spectrum (RSS) method is analyzed quantitatively. It is shown that the split-spectrum processing deteriorates the coherence of the sub-band interferogram and then greatly reduces the estimation accuracy. The RSS method combined with common band filtering (CBF) can improve the coherence of sub-band interferograms and estimation accuracy, but the estimation is biased due to the RSS model mismatch. To address the problem, a modified truncated singular value decomposition (MTSVD) based multi-sub-band RSS method is proposed in this paper. The proposed method divides the range common spectrum into multiple sub-bands to jointly estimate the ionospheric phase. The performance of the proposed method is analyzed and validated based on simulation experiments. The results show that the proposed method has stronger robustness and higher accuracy. [ABSTRACT FROM AUTHOR]

Remote sensing thematic data products are critical for assessing and analyzing geological environments, while efficient generation of thematic products is also highly significant for achieving corresponding sustainable development goals (SDGs). Currently, remote sensing thematic product generation has problems like low levels of automation and efficiency. Addressing these challenges is imperative for advancing sustainable development within the geological environment. This paper aims to address issues related to the generation of geological environment remote sensing thematic products, sorting through the overall process of remote sensing thematic product generation, exploring algorithm encapsulation, combination, and execution under technical methods for container and workflow, and relies on the Spark distributed processing architecture to achieve efficient thematic product generation supported by multiple geological environment data processing models. Finally, taking the three SDGs of SDG6, SDG11, and SDG15 as examples, we achieved the generation of a variety of thematic products such as the interpretation of water body distribution, extraction of urban informal settlements and distribution of water and soil erosion. Meanwhile, we comparatively analyzed the efficiency of thematic product generation on different processing architectures, and the experimental results further verified the feasibility and effectiveness of our proposed solution. This research provides a programme for the automated and intelligent generation of geological environment remote sensing thematic products and effectively assists the construction of sustainable development in the geological environment. [ABSTRACT FROM AUTHOR]

The cross-dimensional matching of 2D images and 3D point clouds is an effective method by which to establish the spatial relationship between 2D and 3D space, which has potential applications in remote sensing and artificial intelligence (AI). In this paper, we propose a novel multi-task network, 2D3D-DescNet, to learn 2D and 3D local feature descriptors jointly and perform cross-dimensional matching of 2D image patches and 3D point cloud volumes. The 2D3D-DescNet contains two branches with which to learn 2D and 3D feature descriptors, respectively, and utilizes a shared decoder to generate the feature maps of 2D image patches and 3D point cloud volumes. Specifically, the generative adversarial network (GAN) strategy is embedded to distinguish the source of the generated feature maps, thereby facilitating the use of the learned 2D and 3D local feature descriptors for cross-dimensional retrieval. Meanwhile, a metric network is embedded to compute the similarity between the learned 2D and 3D local feature descriptors. Finally, we construct a 2D-3D consistent loss function to optimize the 2D3D-DescNet. In this paper, the cross-dimensional matching of 2D images and 3D point clouds is explored with the small object of the 3Dmatch dataset. Experimental results demonstrate that the 2D and 3D local feature descriptors jointly learned by 2D3D-DescNet are similar. In addition, in terms of 2D and 3D cross-dimensional retrieval and matching between 2D image patches and 3D point cloud volumes, the proposed 2D3D-DescNet significantly outperforms the current state-of-the-art approaches based on jointly learning 2D and 3D feature descriptors; the cross-dimensional retrieval at TOP1 on the 3DMatch dataset is improved by over 12%. [ABSTRACT FROM AUTHOR]

Near-range radar imaging (NRRI) has evolved into a vital technology with diverse applications spanning fields such as remote sensing, surveillance, medical imaging and non-destructive testing. The Pseudopolar Format Matrix (PFM) has emerged as a promising technique for representing radar data in a compact and efficient manner. In this paper, we present a comprehensive PFM description of near-range radar imaging. Furthermore, this paper also explores the integration of the Fractional Fourier Transform (FrFT) with PFM for enhanced radar signal analysis. The FrFT—a powerful mathematical tool for signal processing—offers unique capabilities in analysing signals with time-frequency localization properties. By combining FrFT with PFM, we have achieved significant advancements in radar imaging, particularly in dealing with complex clutter environments and improving target detection accuracy. Meanwhile, this paper highlights the imaging matrix form of FrFT under the PFM, emphasizing the potential for addressing challenges encountered in near-range radar imaging. Finally, numerical simulation and real-world scenario measurement imaging results verify optimized accuracy and computational efficiency with the fusion of PFM and FrFT techniques, paving the way for further innovations in near-range radar imaging applications. [ABSTRACT FROM AUTHOR]

Large-Scale land cover mapping (LLCM) based on deep learning models necessitates a substantial number of high-precision sample datasets. However, the limited availability of such datasets poses challenges in regularly updating land cover products. A commonly referenced method involves utilizing prior products (PPs) as labels to achieve up-to-date land cover mapping. Nonetheless, the accuracy of PPs at the regional level remains uncertain, and the Remote Sensing Image (RSI) corresponding to the product is not publicly accessible. Consequently, the sample dataset constructed through geographic location matching may lack precision. Errors in such datasets are not only due to inherent product discrepancies, and can also arise from temporal and scale disparities between the RSI and PPs. In order to solve the above problems, this paper proposes an LLCM framework for generating labels for use with PPs. The framework consists of three main parts. First, initial generation of labels, in which the collected PPs are integrated based on D-S evidence theory and initial labels are obtained using the generated trust map. Second, for dynamic label correction, a two-stage training method based on initial labels is adopted. The correction model is pretrained in the first stage, then the confidence probability (CP) correction module of the dynamic threshold value and NDVI correction module are introduced in the second stage. The initial labels are iteratively corrected while the model is trained using the joint correction loss, with the corrected labels obtained after training. Finally, the classification model is trained using the corrected labels. Using the proposed land cover mapping framework, this study used PPs to produce a 10 m spatial resolution land cover map of Cambodia in 2020. The overall accuracy of the land cover map was 91.68% and the Kappa value was 0.8808. Based on these results, the proposed mapping framework can effectively use PPs to update medium-resolution large-scale land cover datasets, and provides a powerful solution for label acquisition in LLCM projects. [ABSTRACT FROM AUTHOR]

The suddenness of landslide disasters often causes significant loss of life and property. Accurate assessment of landslide disaster susceptibility is of great significance in enhancing the ability of accurate disaster prevention. To address the problems of strong subjectivity in the selection of assessment indicators and low efficiency of the assessment process caused by the insufficient application of a priori knowledge in landslide susceptibility assessment, in this paper, we propose a novel landslide susceptibility assessment framework by combing domain knowledge graph and machine learning algorithms. Firstly, we combine unstructured data, extract priori knowledge based on the Unified Structure Generation for Universal Information Extraction Pre-trained model (UIE) fine-tuned with a small amount of labeled data to construct a landslide susceptibility knowledge graph. We use Paired Relation Vectors (PairRE) to characterize the knowledge graph, then construct a target area characterization factor recommendation model by calculating spatial correlation, attribute similarity, Term Frequency–Inverse Document Frequency (TF-IDF) metrics. We select the optimal model and optimal feature combination among six typical machine learning (ML) models to construct interpretable landslide disaster susceptibility assessment mapping. Experimental validation and analysis are carried out on the three gorges area (TGA), and the results show the effectiveness of the feature factors recommended by the knowledge graph characterization learning, with the overall accuracy of the model after adding associated disaster factors reaching 87.2%. The methodology proposed in this research is a better contribution to the knowledge and data-driven assessment of landslide disaster susceptibility. [ABSTRACT FROM AUTHOR]

Floods are among the most common disasters, causing loss of life and enormous damage to private property and public infrastructure. Monitoring systems that detect and predict floods help respond quickly in the pre-disaster phase to prevent and mitigate flood risk and damages. Thus, this paper presents a deep neural network (DNN)-based real-time flood monitoring system for embedded Internet of Things (IoT) edge devices. The proposed system fuses long-wave infrared (LWIR) and RGB cameras to overcome a critical drawback of conventional RGB camera-based systems: severe performance deterioration at night. This system recognizes areas occupied by water using a DNN-based semantic segmentation network, whose input is a combination of RGB and LWIR images. Flood warning levels are predicted based on the water occupancy ratio calculated by the water segmentation result. The warning information is delivered to authorized personnel via a mobile message service. For real-time edge computing, the heavy semantic segmentation network is simplified by removing unimportant channels while maintaining performance by utilizing the network slimming technique. Experiments were conducted based on the dataset acquired from the sensor module with RGB and LWIR cameras installed in a flood-prone area. The results revealed that the proposed system successfully conducts water segmentation and correctly sends flood warning messages in both daytime and nighttime. Furthermore, all of the algorithms in this system were embedded on an embedded IoT edge device with a Qualcomm QCS610 System on Chip (SoC) and operated in real time. [ABSTRACT FROM AUTHOR]

Change detection is an essential and widely utilized task in remote sensing that aims to detect and analyze changes occurring in the same geographical area over time, which has broad applications in urban development, agricultural surveys, and land cover monitoring. Detecting changes in remote sensing images is a complex challenge due to various factors, including variations in image quality, noise, registration errors, illumination changes, complex landscapes, and spatial heterogeneity. In recent years, deep learning has emerged as a powerful tool for feature extraction and addressing these challenges. Its versatility has resulted in its widespread adoption for numerous image-processing tasks. This paper presents a comprehensive survey of significant advancements in change detection for remote sensing images over the past decade. We first introduce some preliminary knowledge for the change detection task, such as problem definition, datasets, evaluation metrics, and transformer basics, as well as provide a detailed taxonomy of existing algorithms from three different perspectives: algorithm granularity, supervision modes, and frameworks in the Methodology section. This survey enables readers to gain systematic knowledge of change detection tasks from various angles. We then summarize the state-of-the-art performance on several dominant change detection datasets, providing insights into the strengths and limitations of existing algorithms. Based on our survey, some future research directions for change detection in remote sensing are well identified. This survey paper sheds some light the topic for the community and will inspire further research efforts in the change detection task. [ABSTRACT FROM AUTHOR]

The positioning performance of the Global Navigation Satellite System (GNSS) network real-time kinematic (NRTK) depends on regional atmospheric error modeling. Under normal atmospheric conditions, NRTK positioning provides high accuracy and rapid initialization. However, fluctuations in atmospheric conditions can lead to poor atmospheric error modeling, resulting in significant atmospheric biases that affect the positioning accuracy, initialization speed, and reliability of NRTK positioning. Consequently, this decreases the efficiency of NRTK operations. In response to these challenges, this paper proposes a fast and reliable NRTK positioning method based on sequential ambiguity resolution (SAR) of multi-frequency combined observations. This method processes observations from extra-wide-lane (EWL), wide-lane (WL), and narrow-lane (NL) measurements; performs sequential AR using the LAMBDA algorithm; and subsequently constrains other parameters using fixed ambiguities. Ultimately, this method achieves high precision, rapid initialization, and reliable positioning. Experimental analysis was conducted using Continuous Operating Reference Station (CORS) data, with baseline lengths ranging from 88 km to 110 km. The results showed that the proposed algorithm offers positioning accuracy comparable to conventional algorithms in conventional NRTK positioning and has higher fixed rate and positioning accuracy in single-epoch positioning. On two datasets, the proposed algorithm demonstrated over 30% improvement in time to first fix (TTFF) compared to conventional algorithms. It provides higher precision in suboptimal positioning solutions when conventional NRTK algorithms fail to achieve fixed solutions during the initialization phase. These experiments highlight the advantages of the proposed algorithm in terms of initialization speed and positioning reliability. [ABSTRACT FROM AUTHOR]

The presence of side-lobe noise degrades the image quality and adversely affects the performance of inverse synthetic aperture radar (ISAR) image understanding applications, such as automatic target recognition (ATR), target detection, etc. However, methods reliant on data processing, such as windowing, inevitably encounter resolution reduction, and current deep learning approaches under-appreciate the sparsity inherent in ISAR images. Taking the above analysis into consideration, a convolutional neural network-based process for ISAR side-lobe noise training is proposed in this paper. The proposed processing, based on the ISAR images sparsity characteristic analysis, undergoes enhancements in three core ideas, dataset construction, prior network design, and loss function improvements. In the realm of dataset construction, we introduce a bin-by-bin fast clean algorithm and accelerate the processing speed significantly on the basis of image complete information. Subsequently, a spatial attention layer is incorporated into the prior network designed to augment the network's focus on the crucial regions of ISAR images. In addition, a loss function featuring a weighting factor is devised to ensure the precise recovery of the strong scattering point. Simulation experiments demonstrate that the proposed process achieves significant improvements in both quantitative and qualitative results over the classical denoise methods. [ABSTRACT FROM AUTHOR]

Currently, network real-time kinematic (NRTK) technology is one of the primary approaches used to achieve real-time dynamic high-precision positioning, and virtual reference station (VRS) technology, with its high accuracy and compatibility, has become the most important type of network RTK solution. The key to its successful implementation lies in correctly fixing integer ambiguities and extracting spatially correlated errors. This paper first introduces real-time data processing flow on the VRS server side. Subsequently, an improved ionosphere-weighted VRS approach is proposed based on single-differenced observations of GPS, GAL, and BDS. With the prerequisite of ensuring estimable integer properties of ambiguities, it directly estimates the single-differenced ionospheric delay and tropospheric delay between reference stations, reducing the double-differenced (DD) observation noise introduced by conventional models and accelerating the system initialization speed. Based on this, we provide an equation for generating virtual observations directly based on single-differenced atmospheric corrections without specifying the pivot satellite. This further simplifies the calculation process and enhances the efficiency of the solution. Using Australian CORS data for testing and analysis, and employing the approach proposed in this paper, the average initialization time on the server side was 40 epochs, and the average number of available satellites reached 23 (with an elevation greater than 20°). Two positioning modes, 'Continuous' (CONT) and 'Instantaneous' (INST), were employed to evaluate VRS user positioning accuracy, and the distance covered between the user and the master station was between 20 and 50 km. In CONT mode, the average positioning errors in the E/N/U directions were 0.67/0.82/1.98 cm, respectively, with an average success fixed rate of 98.76% (errors in all three directions were within 10 cm). In INST mode, the average positioning errors in the E/N/U directions were 1.29/1.29/2.13 cm, respectively, with an average success fixed rate of 89.56%. The experiments in this study demonstrate that the proposed approach facilitates efficient ambiguity resolution (AR) and atmospheric parameter extraction on the server side, thus enabling users to achieve centimeter-level positioning accuracy instantly. [ABSTRACT FROM AUTHOR]