Search

The launch of large language models (LLMs) like ChatGPT in late 2022 and the anticipated arrival of future GPT-x iterations have marked the beginning of the generative artificial intelligence (GAI) era. We conducted a systematic review of how to integrate LLMs including GPT and other GAI models into geospatial science, based on 293 papers obtained from four databases of academic publications -- Web of Science (WoS), Scopus, SSRN and arXiv -- 26 papers were eventually included for analysis. We statistically outlined the share of domains where LLMs and other GAI models, the type of data that have been used for these models, and the modelling tasks and roles that they play. We also pointed out the challenges and future directions for the next research agenda -- along with which we could better position ourselves in the mainstream of science and the cutting-edge research paradigm as others leverage insights from the growing data deluge. [ABSTRACT FROM AUTHOR]

Vegetation segmentation via point cloud data can provide important information for urban planning and environmental protection. The point cloud dataset is obtained using light detection and ranging (LiDAR) or RGB-D images. Oblique photogrammetry has received little attention as another important source of point cloud data. We present a pointwise annotated oblique photogrammetry point-cloud dataset that contains rich RGB information, texture, and structural features. This dataset contains five regions of Bengbu, China, with more than twenty thousand samples in this paper. Obviously, previous indoor point cloud semantic segmentation models are no longer applicable to oblique photogrammetry point clouds. A random sampling point transformer (RSPT) network is proposed to enhance vegetation segmentation accuracy. The RSPT model offers both efficient and lightweight architecture. In RSPT, random point sampling is utilized to downsample point clouds, and a local feature aggregation module based on self-attention is designed to extract additional representation features. The network also incorporated residual and dense connections (ResiDense) to capture both local and comprehensive features. Compared to state-of-the-art models, RSPT achieves notable improvements. The intersection over union (IoU) metric increased from 96.0% to 96.5%, the F1-score increased from 90.8% to 97.0%, and the overall accuracy (OA) increased from 91.9% to 96.9%. [ABSTRACT FROM AUTHOR]

Urban rail transit (URT) systems play an evident role in shaping city spatial structures; however, the principles and mechanisms behind this influence are not fully understood. This paper reviews research progress on the coupling relationship between URT and urban space, focusing on big data analysis methods and the timeliness and sequence of coupling effects. It highlights the importance of the temporal dimension in coupling analysis. By thoroughly exploiting data value and extracting key elements, big data technology imparts temporal attributes to these elements, exploring their interaction and influence mechanisms over different time sequences. The paper also discusses the potential application of big data to urban planning to support sustainable urban development. Finally, the paper outlines future research directions, including the deepened application of big data to urban spatial analysis and the role of new data sources in understanding and shaping the coupling relationship between URT and urban space. This analysis offers new perspectives and methodologies for urban development and transportation planning. [ABSTRACT FROM AUTHOR]

Remote sensing image building segmentation, which is essential in land use and urban planning, is evolving with advancements in deep learning. Conventional methods using convolutional neural networks face limitations in integrating local and global information and establishing long-range dependencies, resulting in suboptimal segmentation in complex scenarios. This paper proposes LMSwin_PNet, a novel segmentation network that addresses the SwinTransformer encoder's deficiency in local information processing through a local feature extraction module. Additionally, it features a multiscale nonparametric merging attention module to enhance feature-channel correlations. The network also incorporates the pyramid large-kernel convolution module, replacing the traditional 3 x 3 convolution in the decoder with multibranch large-kernel convolution, thereby achieving a large receptive field and detailed information capture. Comparative analyses on three public building datasets demonstrated the model's superior segmentation performance and robustness. The results show that LMSwin_PNet produced outputs closely matching labels, showing its potential for broader application in remote sensing image segmentation tasks. It achieved achieving an IoU of 72.35% on the Massachusetts Building Dataset, 91.30% on the WHU Building Dataset, and 78.99% on the Inria aerial-image building dataset. The source code will be freely available at https://github.com/ziyanpeng/pzy. [ABSTRACT FROM AUTHOR]

Multi-class change detection can make various ground monitoring projects more efficient and convenient. With the development of deep learning, the multi-class change detection methods have introduced Deep Neural Network (DNN) to improve the accuracy and efficiency of traditional methods. The class imbalance in the image will affect the feature extraction effect of DNN. Existing deep learning methods rarely consider the impact of data on DNN. To solve this problem, this paper proposes a class rebalancing algorithm based on data distribution. The algorithm iteratively trains the SSL model, obtains the distribution of classes in the data, then expands the original dataset according to the distribution of classes, and finally trains the baseline SSL model using the expanded dataset. The trained semantic segmentation model is used to detect multi-class changes in two-phase images. This paper is the first time to introduce the image class balancing method in the multi-class change detection task, so a control experiment is designed to verify the effectiveness and superiority of this method for the unbalanced data. The mIoU of the class rebalancing algorithm in this paper reaches 0.4615, which indicates that the proposed method can effectively detect ground changes and accurately distinguish the types of ground changes. [ABSTRACT FROM AUTHOR]

Rapidly estimating post-disaster building damage via high-resolution remote sensing (HRRS) imagery is essential for initial disaster relief. However, the complex appearance of building damage poses challenges for existing methods. Specifically, relying solely on postdisaster images lacks building boundary guidance, while change detection methods using dual-temporal imageries are prone to introducing false changes. To address these issues, this paper presents a novel weakly supervised approach that leverages pre- and post-disaster HRRS images for building damage detection. The contributions of this paper are twofold. Firstly, a unique framework is proposed to utilize dual-temporal images. Precisely, the proposed method initially extracts fine-grained sub-building-level individuals from pre-disaster images by combining a fully convolutional neural network (FCN)-based method with superpixel segmentation. Then, these details serve as cues to effectively guide the detection of damaged building areas on postdisaster images, thereby enhancing accuracy. Secondly, we propose a weakly supervised method that solely relies on labeling building damage based on image patches but can ultimately yield pixel-level building damage results. Experiments conducted using HRRS images captured during the 2010 Haiti earthquake demonstrate that the proposed method outperforms existing methodologies. This effort of this paper will contribute to the sustainable development of cities and human settlements. [ABSTRACT FROM AUTHOR]

Feature matching is a classic challenge in the computer vision field. In this paper, we propose an innovative graph classification method based on neighborhood motion consistency to eliminate erroneous matches. Specifically, we transform the coordinates of feature matching points into vectors on a unified scale. For a given match, we construct a graph centered around the match and incorporating neighboring matches. Node attributes are designed to represent the similarity between the vector of the central node and those of its neighbors. To facilitate this, we develop a lightweight graph attention neural network dedicated to graph property classification, thereby predicting the accuracy of the match under consideration. To effectively train the model, we employ a random cropping strategy to generate a plethora of diverse graphs for classifier training. We evaluate our method on datasets encompassing translational remote sensing data, rotational and scaled remote sensing imagery produced via random cropping, and nonrigid fisheye datasets. Our algorithm demonstrates superior performance to current state-ofthe- art methods. [ABSTRACT FROM AUTHOR]

Deep learning has been extensively utilized in the assessment of building damage after disasters. However, the field of building damage segmentation faces challenges, such as misjudged regions, high network complexity, and long running times. Hence, this paper proposes a two-stage building damage assessment network called the Efficient Channel Attention and Depthwise Separable Convolutional Neural Network (ECADS-CNN). It aims to quickly detect the types of disaster damage in buildings. Deep object segmentation and deep damage classification networks were integrated into a unified building damage detection network. In this study, the efficient channel attention (ECA) module was used to enhance the performance of building semantic segmentation, and a depthwise separable (DS) module was added to the dimension upscaling process. Finally, untrained disaster dataset images were used to test the robustness of the proposed model by comparing the evaluation results of each disaster. The experiments involve testing a total of five common deep learning models, and the results indicate that the ECADS-CNN model improves the speed by 7.4% and the overall F1 score by 5.2% compared with the baseline model. The comprehensive performance is better than that of mainstream deep learning models. [ABSTRACT FROM AUTHOR]

The increasing availability of global observational data has sparked a demand for deep learning algorithms on spherical grids to enable intelligent analysis at a global scale. However, a spherical surface cannot be subdivided into completely identical grid cells through recursive division, and its nonuniformity and irregular deformations lead to uncertainties in the spherical convolutional neural network (SCNN). This paper proposes a multimetric evaluation method to assess the impact of the icosahedral diamond grid quality on the performance of the SCNN by introducing the random forest algorithm to establish nonlinear relationships between multiple grid quality metrics and the SCNN performance and using feature importance analysis to assign impact weights to each grid quality metric considering the SCNN performance. The results show an R2 score of 0.80 for the evaluation method, with four indicators having different weights: cell wall midpoint ratio (0.47), distance between grid points and neighbouring points (0.29), zone standardized compactness (0.13), and angle between a grid point and its two neighbours (0.11). The cell wall midpoint ratio indicator has the most significant impact on the SCNN performance among all grid indicators. [ABSTRACT FROM AUTHOR]

Geographic simulation models can be used to explore and better understand the geographical environment. Recent advances in geographic and socio-environmental research have led to a dramatic increase in the number of models used for this purpose. Some model repositories provide opportunities for users to explore and apply models, but few provide a general evaluation method for assessing the applicability and recognition of models. In this study, an academic impact evaluation method for models is proposed. Five indices are designed based on their pertinence. The analytical hierarchy process is used to calculate the index weights, and the academic impacts of models are quantified with the weighted sum method. The time range is controlled to evaluate the life-term and annual academic impacts of the models. Some models that met the evaluation criteria from different domains are then evaluated. The results show that the academic impact of a model can be quantified with the proposed method, and the major research areas that models impact are identified. [ABSTRACT FROM AUTHOR]

In recent years, RS and GIS technologies have played an increasingly important role in various aspects of rainfall induced landslide research. In order to systematically understand their application situation, this paper extensively used various visualization analysis technologies for in-depth analysis of 1,161 documents collected by the WOS data platform in the past 27 years by combining quantitative and qualitative methods. Then, this article focuses on sub domain analysis from four aspects: landslide detection and monitoring, prediction models, sensitivity mapping, and risk assessment. The study found that the number of literature in this field has steadily increased and is expected to continue to rise. This literature review has attracted widespread attention from the academic community, but it is challenging to meet research needs. Frequent and effective cooperation between countries, institutions, and authors is very beneficial for promoting progress in this field. The future development direction is a new intelligent hybrid model that integrates multiple research methods. This study can provide researchers in this field with the core research force, hot topic evolution, and future development trends of future rainfall-induced landslides and contribute to landslide prevention and control decision-making and achieving the United Nations'sustainable development goals. [ABSTRACT FROM AUTHOR]

The development of unmanned aerial vehicle (UAV) oblique photogrammetric technology provides a good foundation for the rapid construction of large-scale and high-definition real-scene 3D models. However, due to the limitations of the modeling process, irrelevant feature data cannot be eliminated in the modeling stage. The built models contain irrelevant features and model distortions caused by errors. At present, most existing clipping algorithms cannot effectively clip real-scene 3D models that are organized as a whole or with levels of detail (LODs). Therefore, this paper proposes a novel algorithm for clipping real-scene 3D models from any perspective based on clipping boundary lines that fit the surfaces of the models. The results of the clipping experiments for 3D models constructed with oblique UAV images show that this algorithm can effectively clip any part of the 3D models, that the clipping results of each level model closely fit the corresponding clipping boundary lines, and that the accuracy of the clipping results is very high. Additionally, the time complexity of the algorithm is O(n2). In conclusion, the algorithm proposed in this paper provides correct and effective clipping results for real-scene 3D models with LODs that are constructed with photogrammetric or 3D laser scanning data. [ABSTRACT FROM AUTHOR]

Real-time 3D weather radar data processing makes it possible to efficiently simulate meteorological processes in digital Earth and support the assessment of meteorological disasters. The current real-time meteorological operation system can only deal with radar data within 2D space as a flat map and lacks supporting 3D characteristics. Thus, valuable 3D information imbedded in radar data cannot be completely presented to meteorological experts. Due to the large amount of data and high complexity of radar data 3D operation, regular methods are not competent for supporting real-time 3D radar data processing and representation. This study aims to perform radar data 3D operations with high efficiency and instant speed to provide real-time 3D support for the meteorological field. In this paper, a topological framework composed of basic inner topological objects is proposed along with the quadtree structure and LOD architecture, based on which 3D operations on radar data can be conducted in a split second and 3D information can be presented in real time. As the applications of the proposed topological framework, two widely used 3D algorithms in the meteorological field are also implemented in this paper. Finally, a case study verifies the applicability and validity of the proposed topological framework. [ABSTRACT FROM AUTHOR]

The field of architecture, engineering, and construction (AEC) is continually striving to use resources efficiently and manage complex processes. Now more than ever, there is a strong need for digital transformation in AEC. The improvement in the accessibility of consumer-based head-mounted displays (HMD) is attracting different entertainment and research fields to immersive virtual reality (VR) applications. Building Information Modeling (BIM) is known as a promising technology in AEC. The full potential of BIM is not yet employed to empower this field, however, and this could be a result of some barriers still to be surmounted by BIM in both technological and management perspectives. One of these barriers is the communication and collaboration between design, construction, operation, and maintenance phases. VR can fill this gap by providing additional capabilities for BIM which either were not available before or were not possible to employ in practical ways. In this paper, we systematically review recent research around the application of VR in BIM and discuss the results using the PRISMA flowchart. We discuss the most commonly used technologies, software, and evaluation methods and the various applications of VR in the reviewed papers. Finally, we extend the discussion by summarizing the potential future work in this area. [ABSTRACT FROM AUTHOR]

Accurately predicting the geometric structure of a building's roof as a vectorized representation from a raster image is a challenging task in building reconstruction. In this paper, we propose an efficient and precise parsing method called Roof-Former, based on a vision Transformer. Our method involves three steps: (1) Image encoder and edge node initialization, (2) Image feature fusion with an enhanced segmentation refinement branch, and (3) Edge filtering and structural reasoning. Our method outperforms previous works on the vectorizing world building dataset and the Enschede dataset, with vertex and edge heat map F1-scores increasing from 87.1%, 76.2% to 89.1%, 78.1%, and from 69.7%, 68.8% to 71.2%, 69.5%, respectively. Furthermore, our method demonstrates superior performance compared to the current state-of-the-art based on qualitative evaluations, indicating its effectiveness in extracting global image information while maintaining the consistency and topological validity of the roof structure. [ABSTRACT FROM AUTHOR]

The emergence of the Digital Twin of Earth (DTE) signifies a pivotal advancement in the field of the Digital Earth, which includes observations, simulations, and predictions regarding the state of the Earth system and its temporal evolution. While the Geospatial Data Cubes (GDCs) hold promise in probing the DTE across time, existing GDC approaches are not well-suited for real-time data ingestion and processing. This paper proposes the design and implementation of a real-time ingestion and processing approach in Geospatial Data Cube (RTGDC), achieved by ingesting real-time observation streams into data cubes. The methodology employs a publish/subscribe model for efficient observation ingestion. To optimize observation processing within the data cube, a distributed streaming computing framework is utilized in the implementation of the RTGDC. This approach significantly enhances the real-time capabilities of GDCs, providing an efficient solution that brings the realization of the DTE concept within closer reach. In RTGDC, two cases involving local and global scales were implemented, and their real-time performance was evaluated based on latency, throughput, and parallel efficiency. [ABSTRACT FROM AUTHOR]

With the development of social media, it has become increasingly important to quickly and accurately identify social media texts related to disasters (e.g. typhoon) to aid in rescue and recovery efforts. Currently, multi-class classification and pre-trained language model Bidirectional Encoder Representations from Transformers (BERT) are widely used for text classification. However, most studies on typhoon damage classification are multi-class single-label, which contradicts to the reality that a social media text may correspond to multiple types of damage. Moreover, the outputs of the hidden layers of BERT are not fully utilized. This paper proposes a two-stage multi-class multi-label classification method for typhoon damage assessment by fully integrating the outputs of the hidden layers of BERT. In the first stage, sentence vectors are adopted to identify typhoon damage-related texts. In the second stage, word matrices are applied for multi-class multi-label classification to further classify the texts into five damage categories (i.e. transportation, public, electricity, forestry, and waterlogging). The two stages are trained end-to-end to identify typhoon damage from social media texts. Experiments on SinaWeibo texts during typhoon landfall in Chinese coastal regions demonstrate that the proposed method can effectively improve the accuracy of text classification and comprehensively assess typhoon damage. [ABSTRACT FROM AUTHOR]

Mangroves are one of the important components of Earth's carbon sinks. The current problems of base-model composition strategy of ensemble learning and image features combination are still major challenges in mangrove species classification. This paper constructed two novel adaptive ensemble learning frameworks (AME-EL and AOS-EL) to explored the effect of combing different spatial-resolution optical and SAR images on classification performance, and evaluated the ability in mangrove species classification between dual-polarization and fullpolarization SAR images. Finally, we used the SHAP method to explore the effects of different feature interactions on mangrove species classification. The results indicated that: (1) AME-EL and AOS-EL achieve the fine classification of mangrove species with overall accuracies between 77.50% and 94.77%. (2) Combination of Gaofen-7 multispectral and Gaofen-3 SAR improved the classification accuracy for Kandelia candel, with the F1 score increasing from 26.4% to 40.2%. (3) The VV/VH polarization performed better in the classification, with the F1 scores for Aegiceras corniculatum and Kandelia candel were higher than those of HH/HV and AHV polarization by 7%-16.1% and 5.9%-16.1%, respectively. (4) SAR features interacted well with other spectral features, which made a strong contribution to the classification accuracy of mangrove species, and effectively affect the prediction results. [ABSTRACT FROM AUTHOR]

OpenStreetMap (OSM) has been playing increasingly important roles in location-based services, urban planning, tourism, and disaster management. In this paper, we examine a fundamental issue: can OSM reliably provide its services in the long run? To address this sustainable issue, we propose (1) inner cycle of career stages (i.e. activity levels) for monitoring the sustainable status of OSM and (2) critical mass theory for identifying driving factors that make OSM sustain. These new tools are used to track individual trajectories of editing behavior in 2005-2021. Our results suggest that OSM has been self-sustaining up to now and does not show a sign of decline yet. This is supported by the ever-decreasing turnover time and stabilized turnover rate. Together with the finding that most contributors transitioned to a higher career stage within a month, we identify an accelerated socialization process within OSM. We find further that it is the critical mass of people rather than content that makes OSM sustain. This contrasts with the common wisdom: recruiting more people with small contributions is far more important to the long-term growth of OSM than relying on a group of core members with large contributions. Implications and recommendations are critically discussed. [ABSTRACT FROM AUTHOR]

Fast and high-precision urban scene 3D modeling is the foundational data infrastructure for the digital earth and smart cities. However, due to challenges such as water-area matching difficulties and issues like data redundancy and insufficient observations, existing full-automatic 3D modeling methods often result in water-area missing and many small holes in the models and insufficient local-model accuracy. To overcome these challenges, full-automatic high-precision scene 3D reconstruction method with water-area intelligent complementation on depth maps and mesh optimization is proposed. Firstly, SfM was used to calculated image poses and PatchMatch was used to generated initial depth maps. Secondly, a simplified GAN extracted water-area masks and ray tracing was used achieve high-precision auto-completed water-area depth values. Thirdly, fully connected CRF optimized water-areas and arounds in depth maps. Fourthly, high-precision 3D point clouds were obtained using depth map fusion based on clustering culling and depth least squares. Then, mesh was generated and optimized using similarity measurement and vertex gradients to obtain refined mesh. Finally, highprecision scene 3D models without water-area missing or holes were generated. The results showed that: to compare with the-state-of-art ContextCapture, the proposed method enhances model completeness by 14.3%, raises average accuracy by 14.5% and improves processing efficiency by 63.6%. [ABSTRACT FROM AUTHOR]

It is very important to construct digital twin scenes, which can accurately describe the dynamically changing geographical environment and improve the level of refined management in bridge construction. This article proposes a feature constrained automatic diagnostic analysis method for geometric deformation of bridge digital twins. The geometric deformation feature library of bridge twins was first created to accurately describe structural relationships and behavior characteristics. Secondly, line surface feature constraints were used to extract geometric deformation information from bridge digital twins. Then, a geometric deformation diagnosis algorithm was designed based on an improved Hausdorff method. Finally, a case study was conducted to implement experimental analysis. The experimental results show that the method proposed in this paper can automatically extract the geometric morphology and rapidly calculate line and surface deformations for point cloud bridge digital twins. It achieves an efficiency improvement above 90% and with millimeter-level accuracy, which effectively enhances the diagnostic analysis capabilities for geographical digital twin models. [ABSTRACT FROM AUTHOR]

Road extraction from high-resolution remote sensing images (HRSI) is confronted with the challenge that roads are occluded by other objects, including opaque obstructions and similarly colored areas. This paper proposes a dual convolutional network based on hypergraph and multilevel feature fusion (DHM) for road extraction to address these challenges. The DHM consists of two branch networks (HGNN branch and CNN branch) and a bimodal feature fusion module (BFFM). In the HGNN branch, an algorithm is developed to exploit the shape features of roads and construct hypergraphs on the HRSI. Then, hypergraph neural networks are employed for the first time to capture the longrange context of roads to enhance road connectivity. In the CNN branch, a bottleneck fusion module integrated with an encoderdecoder network structure is built to aggregate multiscale local features. In BFFM, the long-range context from the HGNN branch and the local features from the CNN branch are fused through the designed position converter and enhanced graph reasoning module to achieve the complementary advantages of the dual-branch network. Extensive experiments on three datasets show that DHM outperforms other stateof- the-art methods, especially on the GS-Mountain road dataset. Furthermore, DHM significantly improves road extraction in occluded and similar road regions. [ABSTRACT FROM AUTHOR]

Burned area (BA) plays a pivotal role in fire management and the assessment of fire impact on earth-atmosphere system. Threshold-based segmentation from a single image is an efficient and operational method for detecting BA. However, the great diversity of fire conditions necessitates an adaptive threshold that considers environmental variations. This paper presents a maximum curvature segmentation method to capture the adaptative thresholds. The spectral contrasts in near-infrared (NIR) and shortwave infrared (SWIR) bands were utilized to distinguish BA. The decreased NIR threshold was employed to obtain the burned candidates, and the increased SWIR threshold was then applied to confirm the candidates. Experiments were conducted in different biomes, covering the boreal forest, tropical forest, savanna, and Mediterranean, and different seasons including growing and nongrowing seasons. The thresholds changed in each tile, indicating the algorithm adapted the spatial and temporal variations. Comparison with the Burned Area Reference Database was performed at different biomes, resulting in overall dice coefficient (DC), omission error (OE), commission error (CE), and relative Bias (relB) being 0.86, 0.18, 0.10, and -0.08, respectively. The algorithm provides an avenue for adaptive detection of burned areas, and the single-image based approach can provide real-time burned information for wildfire management systems. [ABSTRACT FROM AUTHOR]

Earth Observation (EO) satellites have produced vast image collections that are freely accessible to society. However, handling these images often surpasses the capabilities of traditional hardware and software for EO data storage and processing, posing challenges for traditional Spatial Data Infrastructure (SDI). To overcome these challenges, innovative cloud computing and distributed systems have been developed, such as matrix databases, MapReduce systems, and web services. These technologies are now integrated into leading-edge platforms, forming a new generation of SDI for big EO data. These platforms have different characteristics in terms of governance, technologies, data access, infrastructure abstractions, data processing, and flexibility to extend their functionality. Our work contributes to the area of SDI for big EO data by proposing a server-side data-processing tool called Brazil Data Cube Workflow Engine (BDC-WE), based on workflow orchestration approach. BDC-WE provides a high-level interface using the openEO API for big EO data accessing and processing, allowing SDI maintainers to easily describe sequences of processes and integrate new algorithms. The architecture proposed in this study was implemented and the prototype was evaluated in two case studies described in this paper. [ABSTRACT FROM AUTHOR]

Remote sensing (RS) images enable high-resolution information collection from complex ground objects and are increasingly utilized in the earth observation research. Recently, RS technologies are continuously enhanced by various characterized platforms and sensors. Simultaneously, artificial intelligence vision algorithms are also developing vigorously and playing a significant role in RS image analysis. In particular, aiming to divide images into different ground elements with specific semantic labels, RS image segmentation could realize the visual acquisition and interpretation. As one of the pioneering methods with the advantages of deep feature extraction ability, deep learning (DL) algorithms have been exploited and proved to be highly beneficial for precise segmentation in recent years. In this paper, a comprehensive review is performed on remote sensing survey systems and various kinds of specially designed deep learning architectures. Meanwhile, DL-based segmentation methods applied on four domains are also illustrated, including geography, precision agriculture, hydrology, and environmental protection issues. In the end, the existing challenges and promising research directions in RS image segmentation are discussed. It is envisioned that this review is able to provide a comprehensive and technical reference, deployment and successful exploitation of DL empowered RS image segmentation approaches. [ABSTRACT FROM AUTHOR]

Multiple tasks within the field of geographical information retrieval and geographical information sciences necessitate toponym matching, which involves the challenge of aligning toponyms that share a common referent. The multiple string similarity approaches struggle when confronted with the complexities associated with unofficial and/ or historical variants of identical toponyms. Also, current state-of-the-art approaches/tools to supervised machine learning rely on labeled samples, and they do not adequately address the intricacies of character replacements either from transliterations or historical shifts in linguistic and cultural norms. To address these issues, this paper proposes a novel matching approach that leverages a deep neural network model empowered by geographic language representation model, known as GeoBERT, which stands for geographic Bidirectional Encoder Representations from Transformers (BERT). This model harnesses the groundbreaking capabilities of the GeoBERT framework by extending a generalized Enhanced Sequential Inference Model architecture and integrating multiple features to enhance the accuracy and robustness of the toponym matching. We present a comprehensive evaluation of the proposed method's performance using three extensive datasets. The findings clearly illustrate that our approach outperforms the individual similarity metrics used in previous studies. [ABSTRACT FROM AUTHOR]

Communication towers provide reliable wireless communication services and play a vital role in modern society. This paper presents an automatic 3D reconstruction framework for communication towers based on point clouds. The antenna is reconstructed by a model-driven method, which constructs an energy function including the data and smoothing terms to evaluate the model accuracy. The parameters of the antenna model are obtained by minimizing energy function with hybrid Genetic Algorithm (GA) and Simulated Annealing (SA). And the predefined model is transformed using these parameters to achieve antenna 3D reconstruction. For tower body reconstruction, the point cloud of the tower body is finely extracted with the region growing followed by piecewise fitting to obtain the model. The performance of the proposed method is evaluated on point clouds from six scenes. The average root-mean-square error (RMSE) of antenna reconstruction is 1.3 cm. The average precision, recall, and F1-score values of the tower body point extraction are 99.9%, 93.5%, and 96.6%, respectively. The average RMSE of tower body reconstruction is 1.2 cm. The experimental results indicate that the method can effectively reconstruct the antenna and tower body with point clouds, providing a feasible framework for the 3D reconstruction of communication towers. [ABSTRACT FROM AUTHOR]

Simultaneous Localization and Mapping (SLAM) has been widely used in emergency response, self-driving and city-scale 3D mapping and navigation. Recent deep-learning based feature point extractors have demonstrated superior performance in dealing with the complex environmental challenges (e.g. extreme lighting) while the traditional extractors are struggling. In this paper, we have successfully improved the robustness and accuracy of a monocular visual SLAM system under various complex scenes by adding a deep learning based visual localization thread as an augmentation to the visual SLAM framework. In this thread, our feature extractor with an efficient lightweight deep neural network is used for absolute pose and scale estimation in real time using the highly accurate georeferenced prior map database at 20cm geometric accuracy created by our in-house and low-cost LiDAR and camera integrated device. The closed-loop error provided by our SLAM system with and without this enhancement is 1.03m and 18.28m respectively. The scale estimation of the monocular visual SLAM is also significantly improved (0.01 versus 0.98). In addition, a novel camera-LiDAR calibration workflow is also provided for large-scale 3D mapping. This paper demonstrates the application and research potential of deep-learning based vision SLAM with image and LiDAR sensors. [ABSTRACT FROM AUTHOR]

Numerous news or event pictures are taken and shared on the internet every day that have abundant information worth being mined, but only a small fraction of them are geotagged. The visual content of the news image hints at clues of the geographical location because they are usually taken at the site of the incident, which provides a prerequisite for geo-localization. This paper proposes an automated pipeline based on deep learning for the geo-localization of news pictures in a large-scale urban environment using geotagged street view images as a reference dataset. The approach obtains location information by constructing an attention-based feature extraction network. Then, the image features are aggregated, and the candidate street view image results are retrieved by the selective matching kernel function. Finally, the coordinates of the news images are estimated by the kernel density prediction method. The pipeline is tested in the news pictures in Hong Kong. In the comparison experiments, the proposed pipeline shows stable performance and generalizability in the large-scale urban environment. In addition, the performance analysis of components in the pipeline shows the ability to recognize localization features of partial areas in pictures and the effectiveness of the proposed solution in news picture geo-localization. [ABSTRACT FROM AUTHOR]

Location-Based Services (LBSs) are essential in many application contexts like ride-sharing or navigation apps. There are cases where users could gain an advantage by submitting fake locations. The problem faced in this paper concerns the possibility that the geographic location declared by a user is not the actual location in which the user is placed. Some solutions are based on centralized or distributed verification in the literature, and other solutions are based on witnesses or infrastructure. In this paper, we highlight the limitations of such approaches and propose a new scheme that exploits signals coming from satellites to provide trustworthy location proofs, also respecting users' privacy. The proposed approach is decentralized because location proofs are stored by users in a suitably-encrypted way, and a blockchain is adopted to guarantee data integrity and authenticity. We show that the proposed approach overcomes the state of the art through a detailed analysis. [ABSTRACT FROM AUTHOR]

Supporting distributed real-time collaborative work has become an important development trend in GIS. It enables group users to cooperate and improve the use of scientific data and collaboration technologies. Real-time collaborative GIS (RCGIS) provides a platform for supporting synchronous collaboration using geospatial data in various forms. Traditional RCGIS generally works on a single map view, which has the disadvantage of collaborative interface confusion. The traditional system-driven mechanism of RCGIS is a computer-interaction event that has the disadvantage of an unfriendly collaborative perception. This paper presents a design and prototype of a RCGIS based on multi-view and geo-event driven mechanisms. The geo-event-driven mechanism provides users with a smoother collaborative perception and interactions that are more natural and friendly. The collaboration process for a GIS that is driven by geo-events is also discussed. A multi-view technique is used to make real-time GIS collaboration more orderly. The paper also proposes a synchronization strategy for public–private views. Finally, walk-through examples are used to demonstrate the use of RCGIS within a web environment. [ABSTRACT FROM AUTHOR]

Although the surface energy budget is essential to determine Earth's climate, site measurements of various radiative components are still too scarce to properly characterize their spatial and temporal variations. This has led to the development of a growing number of surface radiation products, mainly including remotely sensed data, model reanalysis data, and simulations using General Circulation Models (GCMs). This collection of papers introduces new techniques, including the use of machine learning methods for radiation estimation, and evaluates and compares various radiation products, as well as their spatio-temporal variations. These studies show large discrepancies among various products across nearly all radiative parameters in either accuracy or spatio-temporal variations. However, remotely sensed radiation products perform relatively better than others. Despite this, there is an urgent need for further efforts to address these discrepancies and improve the accuracy of these estimates. Even though the major radiative parameters including downward shortwave radiation, net longwave radiation, and albedo, from most products show insignificant long-term variation trends on a global scale, only specific regions, such as the Yunnan-Kweichow Plateau (YKP) and regions with permafrost (i.e. Qinghai-Tibet Plateau and Arctic) and glaciers (i.e. Altai Mountains) exhibit remarkable trends. [ABSTRACT FROM AUTHOR]

The provision of urban green spaces has been widely acknowledged to offer numerous benefits to individuals and communities. Recently, there has been a growing recognition of the importance of evaluating urban greenness from a human-centered perspective, partly due to the rapid development of eye-level greenness measurement. To provide a comprehensive overview of measurement approaches and topics related to eye-level urban greenness, we conducted a systematic review of 77 relevant papers. Our review revealed that street view images are the primarily utilized data source to quantify eye-level urban greenness in urban analysis. Additionally, simulated eye-level urban greenness represents a new frontier for the general public and urban planners, offering the potential to acquire universally applicable quantitative greenness data. Our findings further suggest that eye-level urban greenness has been explored across various domains, including physical behavior, public health, urban planning, human perception, and inequity. Such exploration has highlighted a significant association between eye-level urban greenness and individuals and communities. Our review generates new insights on the integration of urban greenness into urban planning and design, encouraging a greater focus on eye-level urban greenness as a critical component of urban environmental quality. [ABSTRACT FROM AUTHOR]

With increased urbanization and the impacts of climate change, cities around the world are making resilience-building a priority. Simultaneously, advances in technology have enabled the creation of City Digital Twins (CDTs). Informed by a literature review and interviews with resilience and Digital Twin experts, this paper explores how CDTs might support the development of more resilient urban communities. First, the various definitions of urban resilience, smart cities and CDTs are described. Second, the paper explores how characteristics of CDTs make them uniquely equipped to facilitate (1) a better understanding of complex phenomena, (2) the imagination of possible futures and (3) collaboration between stakeholders. Finally, the technical requirements and challenges of CDT implementation are discussed, including (1) identifying priority hazards and users, (2) collecting and managing data, (3) integrating different models and (4) ensuring usability. The paper concludes by emphasizing the important role of stakeholders in shaping CDTs that can be successfully integrated by the communities they serve. [ABSTRACT FROM AUTHOR]

The urban mobility structure is a summary of individual movement patterns and the interaction between persons and the urban environment, which is extremely important for urban management and public transportation route planning. The majority of current research on urban mobility structure discovery utilizes the urban environment as a static network to detect the relationship between people groups and urban areas, ignoring the vital problem of how individuals affect urban mobility structure dynamically. In this paper, we propose a spatio-temporal representational learning method based on reinforcement learning for discovering urban mobility structures, in which the model can effectively consider the interaction knowledge graph of individuals with stations while accounting for the spatio-temporal heterogeneity of individual travel. The experimental results demonstrate the advantages of individual travel-based urban mobility structure discovery research in describing the interaction between individuals and urban areas, which can account for the intrinsic influence more thoroughly. [ABSTRACT FROM AUTHOR]

Extracting precise geographical information from the textual content, referred to as toponym recognition, is fundamental in geographical information retrieval and crucial in a plethora of spatial analyses, e.g. mining location-based information from social media, news reports, and surveys for various applications. However, the performance of existing toponym recognition methods and tools is deficient in supporting tasks that rely on extracting fine-grained geographic information from texts, e.g. locating people sending help requests with addresses through social media during disasters. The emerging pretrained language models have revolutionized natural language processing and understanding by machines, offering a promising pathway to optimize toponym recognition to underpin practical applications. In this paper, TopoBERT, a uniquely designed toponym recognition module based on a one-dimensional Convolutional Neural Network (CNN1D) and Bidirectional Encoder Representation from Transformers (BERT), is proposed and fine-tuned. Three datasets are leveraged to tune the hyperparameters and discover the best strategy to train the model. Another seven datasets are used to evaluate the performance. TopoBERT achieves state-of-the-art performance (average f1-score = 0.854) compared to the seven baseline models. It is encapsulated into easy-to-use python scripts and can be seamlessly applied to diverse toponym recognition tasks without additional training. [ABSTRACT FROM AUTHOR]

This paper develops a social media-disaster resilience analysis framework by categorizing types of social media use and their challenges to better understand and assess its role in disaster resilience research and management. The framework is derived primarily from several case studies of Twitter use in three hurricane events in the United States – Hurricanes Isaac, Sandy, and Harvey. The paper first outlines four major contributions of social media data for disaster resilience research and management, which include serving as an effective communication platform, providing ground truth information for emergency response and rescue operations, providing information on people's sentiments, and allowing predictive modeling. However, there are four key challenges to its uses, which include easy spreading of false information, social and geographical disparities of Twitter use, technical issues on processing and analyzing big and noisy data, especially on improving the locational accuracy of the tweets, and algorithm bias in AI and other types of modeling. Then, the paper proposes twenty strategies that the four sectors of the social media community – organizations, individuals, social media companies, and researchers – could take to improve social media use to increase disaster resilience. [ABSTRACT FROM AUTHOR]

Deep learning algorithms show good prospects for remote sensing flood monitoring. They mostly rely on huge amounts of labeled data. However, there is a lack of available labeled data in actual needs. In this paper, we propose a high-resolution multi-source remote sensing dataset for flood area extraction: GF-FloodNet. GF-FloodNet contains 13388 samples from Gaofen-3 (GF-3) and Gaofen-2 (GF-2) images. We use a multi-level sample selection and interactive annotation strategy based on active learning to construct it. Compare with other flood-related datasets, GF-FloodNet not only has a spatial resolution of up to 1.5 m and provides pixel-level labels, but also consists of multi-source remote sensing data. We thoroughly validate and evaluate the dataset using several deep learning models, including quantitative analysis, qualitative analysis, and validation on large-scale remote sensing data in real scenes. Experimental results reveal that GF-FloodNet has significant advantages by multi-source data. It can support different deep learning models for training to extract flood areas. There should be a potential optimal boundary for model training in any deep learning dataset. The boundary seems close to 4824 samples in GF-FloodNet. We provide GF-FloodNet at and . [ABSTRACT FROM AUTHOR]

Understanding the characteristics of passenger vehicle use is the prerequisite for effective urban management. However, it has been challenging in the existing literature due to the lack of continuously observed data on passenger vehicle use. Thanks to the advances in data collection and processing techniques, multi-day vehicle trajectory data generated from volunteered passenger cars provide new opportunities for examining in depth how people travel in regular patterns. In this paper, based on a week's operation data of 6600 passenger cars in Shanghai, we develop a systematic approach for identifying trips and travel purposes, and classify vehicles into four categories using a Gaussian-Mixed-Model. A new method is proposed to identify vehicle travel regularities and we use the Z Test to explore differences in travel time and route choices between four types of vehicles. We find that commercially used vehicles present high travel intensity in temporal and spatial aspects and the use intensity in elevated roads is higher for household-used commuting vehicles than semi-commercially used vehicles. The methodologies and conclusions of this paper may provide not only theoretical support for future urban traffic prediction, but also guidance for employing customized active traffic demand management measures to alleviate traffic congestion. [ABSTRACT FROM AUTHOR]

The United Nations adopted 17 Sustainable Development Goals (SDGs) to address societal, economic and environmental sustainability issues. The efficiency of SDGs monitoring could be improved by essential variables (EVs), which can help to better deal with massive data, interdisciplinary knowledge and workloads. However, in practice, effectively combining EVs with SDGs monitoring remains challenging. In this paper, we proposed a refining method of essential SDGs variables (ESDGVs) to land degradation. Firstly, we selected northwest China as our experimental region and extracted a group of variables related to land degradation from SDG indicators based on the DPSIR framework. Next, we identify the essential ones using a combined qualitative and quantitative methods with the criteria of feasibility, spatialization, and relevance which considered the issues of data acquisition, monitoring scale, and closeness to the land degradation. Finally, we analysed the monitoring role of ESDGVs. Results show that, compared to conventional observations, ESDGVs facilitate the monitoring and evaluation of regional SDGs with reduced efforts. And both climate and human activities have a facilitating or inhibiting effect on land degradation processes. In the future, we hope to have more mature data sets and consider adding more SDG indicators for ESDGVs' refinement. [ABSTRACT FROM AUTHOR]

The different imaging conditions of high spatial resolution remote sensing images (HSRRSIs) tend to cause large differences in the background information of bridges from the images, including problems of difficult detection of multiscale bridges, leakage of small bridges and insufficient detection accuracy for their detection. To address these problems, a YOLOv5 network with a decoupled head for the automatic detection of bridges in HSRRIs is proposed in this paper. First, the problem of inconsistent scale of information fusion of each feature in the feature pyramid network is solved using a weighted bi-directional feature pyramid network (BiFPN). Then, the convolutional block attention module (CBAM) is fused into the three effective feature layers after feature pyramid network processing. The bridge feature information is effectively extracted from the channel and spatial dimensions. Next, the decoupled head is fused in the YOLO Head to separate the classifier and regressor to speed up the network convergence and improve the network detection accuracy simultaneously. Finally, the practical effect is evaluated by calculating the average precision (AP). According to the experimental results, the AP of the proposed method is 98.1%, which is improved by 4.1%∼23.5% compared with other models. [ABSTRACT FROM AUTHOR]

The anomalous movements of glaciers cause disasters, such as debris flows and landslides. It is very important to assess the glacier movements and their future trends. Glacier velocity refers to movement process. The current research aims to analyse past and current spatiotemporal changes in glacier velocity. No study has used neural network model to conduct a spatiotemporal prediction for glacier velocity. Therefore, this paper selected typical mountain glaciers G2 and G5 along the Sichuan-Tibet Railway as research objects and constructed the Convolutional Gate Recurrent Unit (ConvGRU) spatiotemporal prediction model based on 1988–2018 Landsat data to predict velocities in 2019–2028, and analysed the future trends of G2 and G5. The evaluation indexes met the model requirements to a large extent, quantitatively showing that the model has high accuracy and can successfully capture the fluctuation changes in time series data of glacier velocity. The mean deviations of G2 and G5 were 0.09 and −0.47 m/yr, respectively, reflecting the high reliability of the model applied to extraction of glacier velocity. The velocities of G2 and G5 showed a slow downtrend with fluctuations; that is, they will not cause damage to the construction and operation of the Sichuan-Tibet Railway in the short term. [ABSTRACT FROM AUTHOR]

Floods occur frequently worldwide. The timely, accurate mapping of the flooded areas is an important task. Therefore, an unsupervised approach is proposed for automated flooded area mapping from bi-temporal Sentinel-2 multispectral images in this paper. First, spatial–spectral features of the images before and after the flood are extracted to construct the change magnitude image (CMI). Then, the certain flood pixels and non-flood pixels are obtained by performing uncertainty analysis on the CMI, which are considered reliable classification samples. Next, Generalized Regression Neural Network (GRNN) is used as the core classifier to generate the initial flood map. Finally, an easy-to-implement two-stage post-processing is proposed to reduce the mapping error of the initial flood map, and generate the final flood map. Different from other methods based on machine learning, GRNN is used as the classifier, but the proposed approach is automated and unsupervised because it uses samples automatically generated in uncertainty analysis for model training. Results of comparative experiments in the three sub-regions of the Poyang Lake Basin demonstrate the effectiveness and superiority of the proposed approach. Moreover, its superiority in dealing with uncertain pixels is further proven by comparing the classification accuracy of different methods on uncertain pixels. [ABSTRACT FROM AUTHOR]

Due to the existence of seasonal or perennial sea ice cover, the determination of the Arctic sea surface is more difficult than that of mid-low latitudinal oceans. Focusing on the sea surface height in the ice-covered region, this paper constructs a new Arctic mean sea surface (MSS) model, named SUST22, by combining the measurements from ICESat and Cryosat-2 missions. The lead detection methods of ICESat and Cryosat-2 are first studied and modified to acquire sea surface measurements with better accuracy. The results have shown that the standard deviation of Cryosat-2-derived Arctic sea surface height is about 3–4 cm in 10-km resolution grids, while the value of ICESat is 5–6 cm. Then the MSS construction procedure is discussed and the SUST22 MSS model is constructed. The new model is compared with the other four Arctic MSS models. The best agreement is found between SUST22 and DTU21 with an average difference of −4.0 ± 5.2 cm. These models are also validated by ICESat-2 samples. The average difference between ICESat-2 and SUST22 is 15.8 ± 7.4 cm, which shows that the new model SUST22 presents better consistency with the ICESat-2 than any of the other models. [ABSTRACT FROM AUTHOR]

Exploring urban land use change is a classical problem in urban geography. Taking Zhengzhou as an example, this paper analyzes the spatial and temporal characteristics and driving factors of urban land use change, and simulates the spatial pattern of urban land use in the future. The results of the study show that the land use types in Zhengzhou city were mainly farmland and construction land, the area of forestland, grassland, water area, and unused land was smaller, and the main land use change was the transformation of farmland into construction land. The accuracy check of the simulated land use type data in 2020 showed that the kappa coefficient reached 0.9445, which met the accuracy requirement. Then, according to the predicted land use data in 2025, it was found that the area of grassland, construction land, and water area may have decreased, and the area of farmland, forestland and unused land may have increased. Based on the driving force analysis of land use changes, its prediction results can provide an important reference basis for the formulation and planning of policies related to urban construction. [ABSTRACT FROM AUTHOR]

Anomaly detection in Hyperspectral Imagery (HSI) has received considerable attention because of its potential application in several areas. Numerous anomaly detection algorithms for HSI have been proposed in the literature; however, due to the use of different datasets in previous studies, an extensive performance comparison of these algorithms is missing. In this paper, an overview of the current state of research in hyperspectral anomaly detection is presented by broadly dividing all the previously proposed algorithms into eight different categories. In addition, this paper presents the most comprehensive comparative analysis to-date in hyperspectral anomaly detection by evaluating 22 algorithms on 17 different publicly available datasets. Results indicate that attribute and edge-preserving filtering-based detection (AED), local summation anomaly detection based on collaborative representation and inverse distance weight (LSAD-CR-IDW) and local summation unsupervised nearest regularized subspace with an outlier removal anomaly detector (LSUNRSORAD) perform better as indicated by the mean and median values of area under the receiver operating characteristic (ROC) curves. Finally, this paper studies the effect of various dimensionality reduction techniques on anomaly detection. Results indicate that reducing the number of components to around 20 improves the performance; however, any further decrease deteriorates the performance. [ABSTRACT FROM AUTHOR]

With the development of satellite remote sensing technology, image classification task, as the basis of remote sensing data interpretation, has received wide attention to improving accuracy and robustness. At the same time, in-depth learning technology has been widely used in remote sensing and has a far-reaching impact. Since the existing image classification methods ignore the feature that the general image semantics are the same as the semantics of a single pixel, this paper presents an algorithm that uses the semantics of an image to achieve high-precision image classification. Based on the idea of partial substitution for global, this algorithm designs a split result voting mechanism and builds a Vgg-Vote network model. This mechanism votes on the semantically segmented result of an image and uses the maximum filtering function to select the category containing the most significant number of pixels as the prediction category of the image. Experiments on UC Merced Land-User complete datasets and five types of incomplete datasets with varying degrees of interference, including noise, data occlusion and loss, show that the Vote mechanism dramatically improves the classification accuracy, robustness and anti-jamming capability of Vgg-Vote. [ABSTRACT FROM AUTHOR]

Earth observations, especially satellite data, have produced a wealth of methods and results in meeting global challenges, often presented in unstructured texts such as papers or reports. Accurate extraction of satellite and instrument entities from these unstructured texts can help to link and reuse Earth observation resources. The direct use of an existing dictionary to extract satellite and instrument entities suffers from the problem of poor matching, which leads to low recall. In this study, we present a named entity recognition model to automatically extract satellite and instrument entities from unstructured texts. Due to the lack of manually labeled data, we apply distant supervision to automatically generate labeled training data. Accordingly, we fine-tune the pre-trained language model with early stopping and a weighted cross-entropy loss function. We propose the dictionary-based self-training method to correct the incomplete annotations caused by the distant supervision method. Experiments demonstrate that our method achieves significant improvements in both precision and recall compared to dictionary matching or standard adaptation of pre-trained language models. [ABSTRACT FROM AUTHOR]

Thousands of satellites and instruments are providing very unique long-term, refined, and diverse perception capabilities for the states and changes of the Earth's surface environment. When leveraging Earth Observation (EO) techniques in SDG monitoring in specific regions, an important prerequisite is to evaluate whether EO could meet user requirements in terms of spatial coverage, temporal frequency and observing variables or objects. It is highly expected to have a quantitative model that can not only represent EO capabilities and observation requirements but also evaluate the potential of EO capability to fulfil these requirements. This paper first describes EO capabilities from the satellite's orbit, operation time, spatial resolution, revisit cycle, accessibility and observation relevance level to variables. Observation requirements and priorities are then derived from SDG indicators. Finally, the potential model is established to match EO capabilities and user requirements. Taking SDG 14.1.1 as an example, this model is capable of returning an ordered list of satellites and instruments for users to refer to. This model meets the gap of evaluating EO potential to fulfil SDG monitoring. It could make full use of the available EO capabilities worldwide to meet SDG indicator monitoring requirements. [ABSTRACT FROM AUTHOR]

Although research on the discrete global grid systems (DGGSs) has become an essential issue in the era of big earth data, there is still a gap between the efficiency of current encoding and operation schemes for hexagonal DGGSs and the needs of practical applications. This paper proposes a novel and efficient encoding and operation scheme of an optimized hexagonal quadtree structure (OHQS) based on aperture 4 hexagonal discrete global grid systems by translation transformation. A vector model is established to describe and calculate the aperture 4 hexagonal grid system. This paper also provides two different grid code addition algorithms based on induction and i j k coordinate transformation. We implement the transformation between OHQS codes and geographic coordinates through the i j , i j k and I J K coordinate systems. Compared with existing schemes, the scheme in this paper greatly improves the efficiency of the addition operation, neighborhood retrieval and coordinate transformation, and the coding is more concise than other aperture 4 hexagonal DGGSs. The encoding operation based on the i j k coordinate system is faster than the encoding operation based on the induction and addition table. Spatial modeling based OHQS DGGSs are also provided. A case study with rainstorms demonstrated the availability of this scheme. [ABSTRACT FROM AUTHOR]