Search Results (3,349)

The recently proposed concept of “precursory fingerprint” is a logical consequence of the commonsense statement that seismic structures are unique and that their expected preshock behaviors, including precursory phenomena, are also unique. Our new prediction-related research strategy is conceptually based on the principles of (1) the uniqueness of seismogenic structures, (2) interconnected and interacting geospheres, and (3) non-equivalence of Earth’s surface spots in terms of precursory signal receptivity. The precursory fingerprint of a given seismic structure is a unique assemblage of precursory signals of various natures (seismic, physical, chemical, and biological), detectable in principle by using a system of proper monitoring equipment that consists of a matrix of n sensors placed on the ground at “sensitive” spots identified beforehand and on orbiting satellites. In principle, it is composed of a combination of signals that are emitted by the “responsive sensors”, in addition to the “non-responsive sensors”, coming from the sensor matrix, monitoring as many virtual precursory processes as possible by continuously measuring their relevant parameters. Each measured parameter has a pre-established (by experts) threshold value and an uncertainty interval, discriminating between background and anomalous values that are visualized similarly to traffic light signals (green, yellow, and red). The precursory fingerprint can thus be viewed as a particular configuration of “precursory signals” consisting of anomalous parameter values that are unique and characteristic to the targeted seismogenic structure. Presumably, it is a complex entity that consists of pattern, space, and time components. The “pattern component” is a particular arrangement of the responsive sensors on the master board of the monitoring system yielding anomalous parameter value signals, that can be re-arranged, after a series of experiments, in a spontaneously understandable new pattern. The “space component” is a map position configuration of the signal-detecting sensors, whereas the “time component” is a characteristic time sequence of the anomalous signals including the order, occurrence time before the event, transition time between yellow and red signals, etc. Artificial intelligence using pattern-recognition algorithms can be used to follow, evaluate, and validate the precursory signal assemblage and, finally, to judge, together with an expert board of human operators, its “precursory fingerprint” relevance. Signal interpretation limitations and uncertainties related to dependencies on sensor sensibility, focal depth, and magnitude can be established by completing all three phases (i.e., experimental, validation, and implementation) of the precursory fingerprint-based earthquake prediction research strategy. Full article

This study presents a novel bivariate-output deep learning framework based on LeNet-5 for the simultaneous prediction of urban flood susceptibility and inundation depth in Seoul, South Korea. Unlike previous studies that relied on single-output models, the proposed approach jointly learns classification and regression targets through a shared feature extraction structure, enhancing consistency and generalization. Among six tested architectures, the Le5SD_CBAM model—integrating a Convolutional Block Attention Module (CBAM)—achieved the best performance, with 83% accuracy, an Area Under the ROC Curve (AUC) of 0.91 for flood susceptibility classification, and a mean absolute error (MAE) of 0.12 m and root mean squared error (RMSE) of 0.18 m for depth estimation. The model’s spatial predictions aligned well with hydrological principles and past flood records, accurately identifying low-lying flood-prone zones and capturing localized inundation patterns influenced by infrastructure and micro-topography. Importantly, it detected spatial mismatches between susceptibility and depth, demonstrating the benefit of joint modeling. Variable importance analysis highlighted elevation as the dominant predictor, while distances to roads, rivers, and drainage systems were also key contributors. In contrast, secondary terrain attributes had limited influence, indicating that urban infrastructure has significantly altered natural flood flow dynamics. Although the model lacks dynamic forcings such as rainfall and upstream inflows, it remains a valuable tool for flood risk mapping in data-scarce settings. The bivariate-output framework improves computational efficiency and internal coherence compared to separate single-task models, supporting its integration into urban flood management and planning systems. Full article

Camera tracking plays a pivotal role in augmented reality geographic information systems (AR-GIS) and location-based services (LBS), serving as a crucial component for accurate spatial awareness and navigation. Current learning-based camera tracking techniques, while achieving superior accuracy in pose estimation, often overlook changes in scale. This oversight results in less stable localization performance and challenges in coping with dynamic environments. To address these tasks, we propose a pyramid convolution-based scene coordinate regression network (PSN). Our approach leverages a pyramidal convolutional structure, integrating kernels of varying sizes and depths, alongside grouped convolutions that alleviate computational demands while capturing multi-scale features from the input imagery. Subsequently, the network incorporates a novel randomization strategy, effectively diminishing correlated gradients and markedly bolstering the training process’s efficiency. The culmination lies in a regression layer that maps the 2D pixel coordinates to their corresponding 3D scene coordinates with precision. The experimental outcomes show that our proposed method achieves centimeter-level accuracy in small-scale scenes and decimeter-level accuracy in large-scale scenes after only a few minutes of training. It offers a favorable balance between localization accuracy and efficiency, and effectively supports augmented reality visualization in dynamic environments. Full article

Agriculture plays a crucial role in discussions about environmental challenges because of its ecological footprint and high vulnerability to environmental shocks. To better understand the social and behavioral dynamics among food producers and their perceptions of climate change-related risks, this paper draws on forty-one in-depth, semi-structured interviews with farmers and ranchers in Colorado (USA). Leveraging the concept of social capital, the paper extends the concept analytically in a direction missed by previous research highlighting network structures, such as by focusing on its bonding, bridging, and linking characteristics. Instead, focus centers on the inclusiveness and diversity of values, beliefs, worldviews, and cultural orientations within those networks, arguing that these elements can be just as influential, if not more so in certain instances, than structural qualities. The concept of social capital heterogeneity is introduced to describe a network’s level of diversity and inclusivity. The findings do not question the importance of studying network structures when trying to understand how food producers respond to threats like climate change; an approach that remains useful for explaining social learning, technology adoption, and behavioral change. However, this method misses elements captured through a subjective, interpretivist perspective. With social capital heterogeneity, we can use social capital to explore why farmers and ranchers hold specific values and risk perceptions, peering deeper “within” networks, while tools like quantitative social network analysis software help map their structures from the “outside.” Additionally, social capital heterogeneity provides valuable insights into questions about “effective” agro-environmental governance. The paper concludes by discussing practical implications of the findings and reviewing the limitations of the research design. Full article

Floods pose a substantial risk to human well-being. These risks encompass economic losses, infrastructural damage, disruption of daily life, and potential loss of life. This study presents a state-wide and county-level spatial exposure assessment of the Iowa railway network, emphasizing the resilience and reliability of essential services during such disasters. In the United States, the railway network is vital for the distribution of goods and services. This research specifically targets the railway network in Iowa, a state where the impact of flooding on railways has not been extensively studied. We employ comprehensive GIS analysis to assess the vulnerability of the railway network, bridges, rail crossings, and facilities under 100- and 500-year flood scenarios at the state level. Additionally, we conducted a detailed investigation into the most flood-affected counties, focusing on the susceptibility of railway bridges. Our state-wide analysis reveals that, in a 100-year flood scenario, up to 9% of railroads, 8% of rail crossings, 58% of bridges, and 6% of facilities are impacted. In a 500-year flood scenario, these figures increase to 16%, 14%, 61%, and 13%, respectively. Furthermore, our secondary analysis using flood depth maps indicates that approximately half of the railway bridges in the flood zones of the studied counties could become non-functional in both flood scenarios. These findings are crucial for developing effective disaster risk management plans and strategies, ensuring adequate preparedness for the impacts of flooding on railway infrastructure. Full article

The complex geology of southwestern Mexico results from prolonged interaction between the North American and Farallon plates along an active subduction zone. This process led to crustal growth via oceanic lithosphere consumption, island arc accretion and batholith exhumation, forming great geological features like the Guerrero composite terrane. On the other hand, the Zihuatanejo subterrane, evolved into the Jalisco Block is now bounded by major grabens. Aeromagnetic data from the Mexican Geological Service (1962–2016) were used to map geological structures and contribute to the mineral exploration. Advanced magnetic processing and 3D modeling (VOXI Magnetic Vector Inversion) revealed the Jalisco Block’s complex structure, including Triassic basement, Jurassic–Cretaceous volcanics, and plutonic bodies such as the Puerto Vallarta batholith. Magnetic anomalies are related to intrusive bodies and mineralized zones, notably Peña Colorada (Fe), El Barqueño (Au), and La Huerta. Iron deposits are linked to intrusive volcanic–sedimentary contacts, while gold aligns with intrusive zones and observed magnetic maxima. A notable NW–SE magnetic low at 20 km depth suggests a reactivated back-arc basin and crustal fracture zone. These findings underscore aeromagnetic surveys’ value in both mineral exploration and geological interpretation. Full article

Low-level turbulence is a meteorological hazard that disrupts the operation of airports and is particularly pronounced at Hong Kong International Airport (HKIA), which is impacted by various sources of low-level turbulence (e.g., terrain disrupting wind flow, sea breeze, and thunderstorms). The possibility of using root-mean-square vertical acceleration (RMSVA) data from Automatic Dependent Surveillance–Broadcast (ADS-B) for low-level turbulence monitoring is studied in this paper. Comparisons are performed between RMSVA and Light Detection And Ranging (LIDAR)-based Eddy Dissipation Rate (EDR) maps and the aircraft-based EDR. Moreover, the LIDAR-based EDR map, aircraft EDR, and pilot report for turbulence reporting are compared for two typical cases at HKIA. It was found that the various estimates/reports of turbulence are generally consistent with one another, at least based on the limited sample considered in this paper. However, at very low altitudes close to the touchdown of arrival flights, RMSVA may not be available due to a lack of ADS-B data. With effective quality control and further in-depth study, it will be possible to use RMSVA to monitor low-level turbulence and to alert pilots if turbulence is reported by the pilot of the preceding flight based on RMSVA. The technical details of the various comparisons and the assumptions made are described herein. Full article

The increasingly severe issue of marine debris presents a critical threat to the sustainable development of marine ecosystems. Real-time detection is essential for timely intervention and cleanup. Furthermore, the density of marine debris exhibits significant depth-dependent variation, resulting in degraded detection accuracy. Based on 9625 publicly available underwater images spanning various depths, this study proposes UTNet, a lightweight neural model, to improve the effectiveness of real-time intelligent identification of marine debris through multidimensional optimization. Compared to Faster R-CNN, SSD, and YOLOv5/v8/v11/v12, the UTNet model demonstrates enhanced performance in random image detection, achieving maximum improvements of 3.5% in mAP50 and 9.3% in mAP50-95, while maintaining reduced parameter count and low computational complexity. The UTNet model is further evaluated on underwater videos for real-time debris recognition at varying depths to validate its capability. Results show that the UTNet model exhibits a consistently increasing trend in confidence levels across different depths as detection distance decreases, with peak values of 0.901 at the surface and 0.764 at deep-sea levels. In contrast, the other six models display greater performance fluctuations and fail to maintain detection stability, particularly at intermediate and deep depths, with evident false positives and missed detections. In summary, the lightweight UTNet model developed in this study achieves high detection accuracy and computational efficiency, enabling real-time, high-precision detection of marine debris at varying depths and ultimately benefiting mitigation and cleanup efforts. Full article

Visual body-tracking techniques have seen widespread adoption in applications such as motion analysis, human–machine interaction, tele-robotics and extended reality (XR). These systems typically provide 2D landmark coordinates corresponding to key limb positions. However, to construct a meaningful 3D kinematic model for body joint reconstruction, a mapping must be established between these visual landmarks and the underlying joint parameters of individual body parts. This paper presents a method for constructing a 3D kinematic model of the human hand using calibrated 2D landmark-tracking data augmented with depth information. The proposed approach builds a hierarchical model in which the palm serves as the root coordinate frame, and finger landmarks are used to compute both forward and inverse kinematic solutions. Through step-by-step examples, we demonstrate how measured hand landmark coordinates are used to define the palm reference frame and solve for joint angles for each finger. These solutions are then used in a visualization framework to qualitatively assess the accuracy of the reconstructed hand motion. As a future work, the proposed model offers a foundation for model-based hand kinematic estimation and has utility in scenarios involving occlusion or missing data. In such cases, the hierarchical structure and kinematic solutions can be used as generative priors in an optimization framework to estimate unobserved landmark positions and joint configurations. The novelty of this work lies in its model-based approach using real sensor data, without relying on wearable devices or synthetic assumptions. Although current validation is qualitative, the framework provides a foundation for future robust estimation under occlusion or sensor noise. It may also serve as a generative prior for optimization-based methods and be quantitatively compared with joint measurements from wearable motion-capture systems. Full article

Accurate and timely land monitoring is crucial for addressing global environmental, economic, and societal challenges, including climate change, sustainable development, and disaster mitigation. While single-source remote sensing data offers significant capabilities, inherent limitations such as cloud cover interference (optical), speckle noise (radar), or limited spectral information (LiDAR) often hinder comprehensive and robust characterization of land surfaces. Recent advancements in synergistic harmonization technology for land monitoring, along with enhanced signal processing techniques and the integration of machine learning algorithms, have significantly broadened the scope and depth of geosciences. Therefore, it is essential to summarize the comprehensive applications of synergistic harmonization technology for geosciences, with a particular focus on recent advancements. Most of the existing review papers focus on the application of a single technology in a specific area, highlighting the need for a comprehensive review that integrates synergistic harmonization technology. This review provides a comprehensive review of advancements in land monitoring achieved through the synergistic harmonization of optical, radar, and LiDAR satellite technologies. It details the unique strengths and weaknesses of each sensor type, highlighting how their integration overcomes individual limitations by leveraging complementary information. This review analyzes current data harmonization and preprocessing techniques, various data fusion levels, and the transformative role of machine learning and deep learning algorithms, including emerging foundation models. Key applications across diverse domains such as land cover/land use mapping, change detection, forest monitoring, urban monitoring, agricultural monitoring, and natural hazard assessment are discussed, demonstrating enhanced accuracy and scope. Finally, this review identifies persistent challenges such as technical complexities in data integration, issues with data availability and accessibility, validation hurdles, and the need for standardization. It proposes future research directions focusing on advanced AI, novel fusion techniques, improved data infrastructure, integrated “space–air–ground” systems, and interdisciplinary collaboration to realize the full potential of multi-sensor satellite data for robust and timely land surface monitoring. Supported by deep learning, this synergy will improve our ability to monitor land surface conditions more accurately and reliably. Full article

This paper presents an innovative multi-criteria visual quality control algorithm designed for deployment on cost-effective Edge devices within the Industrial Internet of Things environment. Traditional industrial vision systems are typically associated with high acquisition, implementation, and maintenance costs. The proposed solution addresses the need to reduce these costs while maintaining high defect detection efficiency. The developed algorithm largely eliminates the need for time- and energy-intensive neural network training or retraining, though these capabilities remain optional. Consequently, the reliance on human labor, particularly for tasks such as manual data labeling, has been significantly reduced. The algorithm is optimized to run on low-power computing units typical of budget industrial computers, making it a viable alternative to server- or cloud-based solutions. The system supports flexible integration with existing industrial automation infrastructure, but it can also be deployed at manual workstations. The algorithm’s primary application is to assess the spread quality of thick liquid mold filling; however, its effectiveness has also been demonstrated for 3D printing processes. The proposed hybrid algorithm combines three approaches: (1) the classical SSIM image quality metric, (2) depth image measurement using Intel MiDaS technology combined with analysis of depth map visualizations and histogram analysis, and (3) feature extraction using selected artificial intelligence models based on the OpenCLIP framework and publicly available pretrained models. This combination allows the individual methods to compensate for each other’s limitations, resulting in improved defect detection performance. The use of hybrid metrics in defective sample selection has been shown to yield superior algorithmic performance compared to the application of individual methods independently. Experimental tests confirmed the high effectiveness and practical applicability of the proposed solution, preserving low hardware requirements. Full article

Autonomous UAV navigation in unknown and complex environments remains a core challenge, especially under limited sensing and computing resources. While most methods rely on modular pipelines involving mapping, planning, and control, they often suffer from poor real-time performance, limited adaptability, and high dependency on accurate environment models. Moreover, many deep-learning-based solutions either use RGB images prone to visual noise or optimize only a single objective. In contrast, this paper proposes a unified, model-free vision-based DRL framework that directly maps onboard depth images and UAV state information to continuous navigation commands through a single convolutional policy network. This end-to-end architecture eliminates the need for explicit mapping and modular coordination, significantly improving responsiveness and robustness. A novel multi-objective reward function is designed to jointly optimize path efficiency, safety, and energy consumption, enabling adaptive flight behavior in unknown complex environments. The trained policy demonstrates generalization in diverse simulated scenarios and transfers effectively to real-world UAV flights. Experiments show that our approach achieves stable navigation and low latency. Full article

For fruit harvesting robots, rapid and accurate detection of fruits and picking points is one of the main challenges for their practical deployment. Several fruits typically grow in clusters or bunches, such as grapes, cherry tomatoes, and blueberries. For such clustered fruits, it is desired for them to be picked by bunches instead of individually. This study proposes utilizing a low-cost off-the-shelf RGB-D sensor mounted on the end effector and a lightweight improved YOLOv8-Pose neural network to detect cherry tomato bunches and picking points for robotic harvesting. The problem of occlusion and overlap is alleviated by merging RGB and depth images from the RGB-D sensor. To enhance detection robustness in complex backgrounds and reduce the complexity of the model, the Starblock module from StarNet and the coordinate attention mechanism are incorporated into the YOLOv8-Pose network, termed StarBL-YOLO, to improve the efficiency of feature extraction and reinforce spatial information. Additionally, we replaced the original OKS loss function with the L1 loss function for keypoint loss calculation, which improves the accuracy in picking points localization. The proposed method has been evaluated on a dataset with 843 cherry tomato RGB-D image pairs acquired by a harvesting robot at a commercial greenhouse farm. Experimental results demonstrate that the proposed StarBL-YOLO model achieves a 12% reduction in model parameters compared to the original YOLOv8-Pose while improving detection accuracy for cherry tomato bunches and picking points. Specifically, the model shows significant improvements across all metrics: for computational efficiency, model size (−11.60%) and GFLOPs (−7.23%); for pickable bunch detection, mAP50 (+4.4%) and mAP50-95 (+4.7%); for non-pickable bunch detection, mAP50 (+8.0%) and mAP50-95 (+6.2%); and for picking point detection, mAP50 (+4.3%), mAP50-95 (+4.6%), and RMSE (−23.98%). These results validate that StarBL-YOLO substantially enhances detection accuracy for cherry tomato bunches and picking points while improving computational efficiency, which is valuable for resource-constrained edge-computing deployment for harvesting robots. Full article

Robust perception of road surface conditions is a critical challenge for the safe deployment of autonomous vehicles and the efficient management of transportation infrastructure. This paper introduces a synthetic data-driven deep learning framework designed to address this challenge. We present a large-scale, procedurally generated 3D synthetic dataset created in Blender, featuring a diverse range of road defects—including cracks, potholes, and puddles—alongside crucial road features like manhole covers and patches. Crucially, our dataset provides dense, pixel-perfect annotations for segmentation masks, depth maps, and camera parameters (intrinsic and extrinsic). Our proposed model leverages these rich annotations in a multi-task learning framework that jointly performs road defect segmentation and depth estimation, enabling a comprehensive geometric and semantic understanding of the road environment. A core contribution is a two-stage domain adaptation strategy to bridge the synthetic-to-real gap. First, we employ a modified CycleGAN with a segmentation-aware loss to translate synthetic images into a realistic domain while preserving defect fidelity. Second, during model training, we utilize a dual-discriminator adversarial approach, applying alignment at both the feature and output levels to minimize domain shift. Benchmarking experiments validate our approach, demonstrating high accuracy and computational efficiency. Our model excels in detecting subtle or occluded defects, attributed to an occlusion-aware loss formulation. The proposed system shows significant promise for real-time deployment in autonomous navigation, automated infrastructure assessment and Advanced Driver-Assistance Systems (ADAS). Full article

In this research paper, we have implemented a computer program thanks to the XML language to sense the differences in image color depth by using haptic/tactile devices. With the use of “Bump Map” and tools such as “Autodesk’s 3D Studio Max”, “Adobe Photoshop”, and “Adobe Illustrator”, we were able to obtain the desired results. The haptic devices used for the experiments were the “PHANTOM Touch” and the “PHANTOM Omni R” of “3D Systems”. The programs that were installed and configured properly so as to model the surfaces, run the experiments, and finally achieve the desired goal are “H3D Api”, “Geomagic_OpenHaptics”, and “OpenHaptics_Developer_Edition”. The purpose of this project was to feel different textures, shapes, and objects in images by using a haptic device. The primary objective was to create a system from the ground up to render visuals on the screen and facilitate interaction with them via the haptic device. The main focus of this work is to propose a novel pattern of images that we can classify as different textures so that they can be identified by people with reduced vision. Full article