Explore the vast range of titles available.

Monitoring lake and reservoir water levels is critical for water resource management and flood risk mitigation. We integrate Sentinel‐3A/B and ICESat‐2 altimetry to reconstruct monthly water levels (2018–2023) for 7,433 lakes and reservoirs (>5 km2) across Asia and estimate their storage variations. Reservoirs exhibit a median annual water level change of 0.36 m/yr, far exceeding the 0.05 m/yr observed for lakes, highlighting their dominant role in surface water dynamics. Eight Asian basins flood events reveal that insufficient self‐regulation capacity of lakes is the primary flood trigger, while large reservoirs effectively mitigate flood frequency and intensity through regulation. These findings emphasize the importance of high‐precision satellite altimetry in surface water assessments and the critical role of reservoirs in modulating hydrological extremes under climate change.

Precipitation plays a crucial role in the global hydrological cycle, and its irregular distribution contributes directly to natural hazards such as floods, waterlogging, and droughts. Satellite remote sensing has emerged as an effective tool for global precipitation monitoring. However, accurately estimating hourly precipitation from satellite observations remains a major challenge due to its high spatiotemporal variability. To address this challenge, we propose a novel framework—Geographically constrained multi‐source Fusion Network with cross Attention (GeoFNA)—designed to enhance the accuracy of hourly satellite precipitation estimates. GeoFNA integrates a spatiotemporal convolutional network with cross‐attention mechanisms to effectively capture complex spatiotemporal patterns and nonlinear relationships across multi‐source precipitation data sets and auxiliary variables. To further improve model robustness, geographically associated input constraints and weight constraints are incorporated to account for the skewed distribution and rapid variability of hourly precipitation. Results demonstrated that GeoFNA outperformed three baseline models, achieving significantly higher agreement with in situ measurements. Specifically, GeoFNA increased the Pearson Correlation Coefficient from 0.38 to 0.89 and reduced the Mean Squared Error from 2.39 to 0.50 (mm/h)2 compared to the original satellite precipitation data. Additionally, GeoFNA exhibited strong spatial robustness, underscoring its potential for accurate and reliable quantitative precipitation estimation. These advancements pave the way for improved hydrological modeling and meteorological research.

Identifying permanent water and temporary water in flood disasters efficiently has mainly relied on change detection method from multi-temporal remote sensing imageries, but estimating the water type in flood disaster events from only post-flood remote sensing imageries still remains challenging. Research progress in recent years has demonstrated the excellent potential of multi-source data fusion and deep learning algorithms in improving flood detection, while this field has only been studied initially due to the lack of large-scale labelled remote sensing images of flood events. Here, we present new deep learning algorithms and a multi-source data fusion driven flood inundation mapping approach by leveraging a large-scale publicly available Sen1Flood11 dataset consisting of roughly 4831 labelled Sentinel-1 SAR and Sentinel-2 optical imagery gathered from flood events worldwide in recent years. Specifically, we proposed an automatic segmentation method for surface water, permanent water, and temporary water identification, and all tasks share the same convolutional neural network architecture. We utilize focal loss to deal with the class (water/non-water) imbalance problem. Thorough ablation experiments and analysis confirmed the effectiveness of various proposed designs. In comparison experiments, the method proposed in this paper is superior to other classical models. Our model achieves a mean Intersection over Union (mIoU) of 52.99%, Intersection over Union (IoU) of 52.30%, and Overall Accuracy (OA) of 92.81% on the Sen1Flood11 test set. On the Sen1Flood11 Bolivia test set, our model also achieves very high mIoU (47.88%), IoU (76.74%), and OA (95.59%) and shows good generalization ability.

Prediction of pedestrian crossing behavior is an important issue faced by the realization of autonomous driving. The current research on pedestrian crossing behavior prediction is mainly based on vehicle camera. However, the sight line of vehicle camera may be blocked by other vehicles or the road environment, making it difficult to obtain key information in the scene. Pedestrian crossing behavior prediction based on surveillance video can be used in key road sections or accident-prone areas to provide supplementary information for vehicle decision-making, thereby reducing the risk of accidents. To this end, we propose a pedestrian crossing behavior prediction network for surveillance video. The network integrates pedestrian posture, local context and global context features through a new cross-stacked gated recurrence unit (GRU) structure to achieve accurate prediction of pedestrian crossing behavior. Applied onto the surveillance video dataset from the University of California, Berkeley to predict the pedestrian crossing behavior, our model achieves the best results regarding accuracy, F1 parameter, etc. In addition, we conducted experiments to study the effects of time to prediction and pedestrian speed on the prediction accuracy. This paper proves the feasibility of pedestrian crossing behavior prediction based on surveillance video. It provides a reference for the application of edge computing in the safety guarantee of automatic driving.

Land cover products are an indispensable data source in land surface process research, and their accuracy directly affects the reliability of related research. Due to the differences in factors such as satellite sensors, the temporal–spatial resolution of remote sensing images, and landcover interpretation technologies, various recently released land cover products are inconsistent, and their accuracy is usually insufficient to meet application requirements. This study, therefore, established a fusion and correction method for multi-source landcover products by combining them with landcover statistics from the Food and Agriculture Organization of the United Nations (FAO), introducing a spatial consistency discrimination technique, and applying an improved Dempster-Shafer evidence fusion method. The five countries in Central Asia were used for a method application and verification assessment. The nine products selected (CCI-LC, CGLS, FROM-GLC, GLCNMO, MCD12Q, GFSAD30, PALSAR, GSWD, and GHS-BUILT) were consistent in time and covered the study area. Based on the interpretation of 1437 high-definition image verification areas, the overall accuracy of the fusion landcover result was 85.32%, and the kappa coefficient was 0.80, which was better than that of the existing comprehensive products. The spatial consistency fusion method had the advantage of an improved statistical fitting, with an overall similarity statistic of 0.999. The improved Dempster-Shafer evidence theory fusion method had an accuracy that was 4.86% higher than the spatial consistency method, and the kappa coefficient increased by 0.07. Combining these two methods improved the consistency of the multi-source data fusion and correction method established in this paper and will also provide more reliable basic data for future research in Central Asia.

The uncertainties in quality evaluations of rock mass are embedded in the underlying multi-source data composed by a variety of testing methods and some specialized sensors. To mitigate this issue, a proper method of data-driven computing for quality evaluation of rock mass based on the theory of multi-source data fusion is required. As the theory of multi-source data fusion, Dempster–Shafer (D-S) evidence theory is applied to the quality evaluation of rock mass. As the correlation between different rock mass indices is too large to be ignored, belief reinforcement and Murphy’s average belief theory are introduced to process the multi-source data of rock mass. The proposed method is designed based on RMR14, one of the most widely used quality-evaluating methods for rock mass in the world. To validate the proposed method, the data of rock mass is generated randomly to realize the data fusion based on the proposed method and the conventional D-S theory. The fusion results based on these two methods are compared. The result of the comparison shows the proposed method amplifies the distance between the possibilities at different ratings from 0.0666 to 0.5882, which makes the exact decision more accurate than the other. A case study is carried out in Daxiagu tunnel in China to prove the practical value of the proposed method. The result shows the rock mass rating of the studied section of the tunnel is in level III with the maximum possibility of 0.9838, which agrees with the geological survey report.

Process data plays a vital role in diagnosing fault sources in chemical production. However, such data contain rich process information and are often sensitive, making direct analysis infeasible due to privacy concerns. Although federated learning mitigates data leakage risks, the conventional averaging strategy falls short in achieving high fault identification accuracy, especially under non-independent and identically distributed (non-IID) client data. To overcome this challenge, we propose a personalized federated learning framework, in which a Takagi-Sugeno (T-S) fuzzy fusion rule is designed. Then, the personalized model is constructed through a structured procedure: fuzzification of model parameter distances, definition of fuzzy rules, fuzzy inference, and defuzzification. Moreover, layer-wise fusion is employed to enhance the precision of aggregation. Evaluations on the Tennessee Eastman (TE) process demonstrate that our method achieves superior fault identification accuracy. The results validate the efficacy of the proposed Fuzzy Rule-Based Federated Layer-wise Fusion (FedFZ) framework in industrial fault diagnosis under heterogeneous data distributions.

Accurate tool wear identification is essential for ensuring the continuity of intelligent machining and workpiece quality. To address the challenges of multi-source fusion inefficiency and inadequate feature extraction, this study proposes a novel identification architecture combining physics-guided multi-channel Gramian angular field (PG-MGAF) with a minimalist 14-layer CSA-Kansformer network. Multi-source signals are preprocessed via PG-MGAF to convert 1D time-series into 2D RGB images, effectively characterizing spatial coupling and interactive energy across three channels. Subsequently, the minimalist network maps these composite features to tool states, significantly reducing computational overhead. Experimental results demonstrate that the proposed model achieves an average accuracy of 93.6% with a single-step inference latency of only 5.90 ms, significantly outperforming mainstream methods such as MobileNet-V2 and ConvNeXt. This architecture provides a high-efficiency, low-latency solution for real-time tool condition monitoring under complex industrial conditions.

With the increasingly widespread application of UAV intelligence, the need for autonomous navigation and positioning is becoming more and more important. To solve the problem that UAV cannot perform localization in complex scenes, a new multi-source fusion framework factor graph optimization algorithm is used for UAV localization state estimation in this paper, which is based on IMU/GNSS/VO multi-source sensors. Based on the factor graph model and the iSAM incremental inference algorithm, a multi-source fusion model of IMU/GNSS/VO is established, including the IMU pre-integration factor, IMU bias factor, GNSS factor, and VO factor. Mathematical simulations and validations on the EuRoC dataset show that, when the selected sliding window size is 30, the factor graph optimization (FGO) algorithm can not only meet the requirements of real time and accuracy at the same time, but it also achieves a plug-and-play function in the event of local sensor failures. Finally, compared with the traditional federated Kalman algorithm and the adaptive federated Kalman algorithm, the positioning accuracy of the FGO algorithm in this paper is improved by 1.5–2-fold, and can effectively improve autonomous navigation system robustness and flexibility in complex scenarios. Moreover, the multi-source fusion framework in this paper is a general algorithm framework that can satisfy other scenarios and other types of sensor combinations.

This paper addresses three critical challenges in urban traffic digital twin systems: (1) achieving high-fidelity multi-source data fusion, (2) overcoming computational bottlenecks in large-scale real-time traffic simulation, and (3) enabling intuitive dynamic interaction for enhanced decision-making. To this end, we propose and implement a novel Unity-based digital twin system with a three-layer architecture. This architecture comprises: (1) a System Construction Layer integrating Google Maps, BlenderGIS, and CityEngine via Building Information Modeling (BIM) and Geographic Information Systems (GIS) fusion to generate sub-meter accuracy 3D models (Root Mean Square Error (RMSE) ≤ 0.15m) with parametric Computer Generated Architecture (CGA) road editing; (2) a Data Acquisition Layer synchronizing Amap Application Programming Interface (API) traffic flow and OpenWeatherMap weather data to drive real-time environmental responses (e.g., rain particles = API intensity × 80); and (3) a Concept Generation Layer implementing an optimized 3-Degrees of Freedom (3-DoF) vehicle dynamics model with linear tire stiffness. Key innovations include Graphics Processing Unit (GPU)-accelerated collision detection (Unity Physics), an adaptive Level of Detail (LOD) strategy for 1,500-vehicle concurrent simulation at 60 Frames Per Second (FPS), and closed-loop decision feedback. Validated on an i7-13700H/RTX 4060 platform, the system reduces Central Processing Unit (CPU)/GPU utilization by 47%/48% versus nonlinear models while maintaining trajectory error < 0.23m (9.5%) in 80 km/h emergency scenarios. Comprehensive comparative experiments confirm its efficacy, providing crucial technical support for smart traffic management, autonomous driving testing, and policy pre-evaluation.