Search Results (3,043)

With the continuous advancement of deep learning algorithms and the rapid growth of computational resources, deep learning technology has undergone numerous milestone developments, evolving from simple BP neural networks into more complex and powerful network models such as CNNs, LSTMs, RNNs, and GANs. In recent years, the application of deep learning technology in ionospheric modeling has achieved breakthrough advancements, significantly impacting navigation, communication, and space weather forecasting. Nevertheless, due to limitations in observational networks and the dynamic complexity of the ionosphere, deep learning-based ionospheric models still face challenges in terms of accuracy, resolution, and interpretability. This paper systematically reviews the development of deep learning applications in ionospheric modeling, summarizing findings that demonstrate how integrating multi-source data and employing multi-model ensemble strategies has substantially improved the stability of spatiotemporal predictions, especially in handling complex space weather events. Additionally, this study explores the potential of deep learning in ionospheric modeling for the early warning of geological hazards such as earthquakes, volcanic eruptions, and tsunamis, offering new insights for constructing ionospheric-geological activity warning models. Looking ahead, research will focus on developing hybrid models that integrate physical modeling with deep learning, exploring adaptive learning algorithms and multi-modal data fusion techniques to enhance long-term predictive capabilities, particularly in addressing the impact of climate change on the ionosphere. Overall, deep learning provides a powerful tool for ionospheric modeling and indicates promising prospects for its application in early warning systems and future research. Full article

With the continuous progress of remote sensing technology, the demand for means of detecting small targets in remote sensing images is escalating. The significance of detecting small targets in remote sensing images lies in enhancing the ability to identify small and elusive targets and the detection accuracy against complex backgrounds, holding significant application value in military reconnaissance, environmental monitoring, and disaster early-warning systems. Firstly, the minuteness of certain targets in relation to the entire image in which they occur, particularly when the camera is situated at a higher altitude, renders them difficult to detect. Secondly, the varying background and lighting conditions in remote sensing images further complicate the detection task. Conventional target detection methods are frequently incapable of addressing these complexities, resulting in a reduction in detection accuracy and an increase in false alarms. Hence, in this paper, we propose a lightweight remote-sensing image target detection network model, CBGS-YOLO, created by introducing the Ghost module to decrease the model parameters, applying the SPD-Conv module to optimize downsampling, and integrating the convolutional block attention module to enhance detection accuracy. The experimental outcomes demonstrate that CBGS-YOLO outperforms other models when applied to the DB_Licenta and USOD datasets, significantly enhancing detection performance for small targets. Compared with YOLOv9, this model can reduce the number of parameters from 7.10 M to 5.12 M, and the average precision (mAP) is effectively improved. The model strengthens the ability to identify small targets against complex backgrounds while maintaining lightweight properties and possesses remarkable application prospects and practical value. Full article

Floods have catastrophic effects worldwide, particularly in monsoonal Asia. This systematic review investigates the literature from the past two decades, focusing on the use of remote sensing (RS), Geographic Information Systems (GISs), and technologies for flood disaster management in South Asia, and addresses the urgent need for effective strategies in the face of escalating flood disasters. This study emphasizes the importance of tailored GIS- and RS-based flood disaster studies inspired by diverse research, particularly in India, Pakistan, Bangladesh, Sri Lanka, Nepal, Bhutan, Afghanistan, and the Maldives. Our dataset comprises 94 research articles from Google Scholar, Scopus, and ScienceDirect. The analysis revealed an upward trend after 2014, with a peak in 2023 for publications on flood-related topics, primarily within the scope of RS and GIS, flood-risk monitoring, and flood-risk assessment. Keyword analysis using VOSviewer revealed that out of 6402, the most used keyword was “climate change”, with 360 occurrences. Bibliometric analysis shows that 1104 authors from 52 countries meet the five minimum document requirements. Indian and Pakistani researchers published the most number of papers, whereas Elsevier, Springer, and MDPI were the three largest publishers. Thematic analysis has identified several major research areas, including flood risk assessment, flood monitoring, early flood warning, RS and GIS, hydrological modeling, and urban planning. RS and GIS technologies have been shown to have transformative effects on early detection, accurate mapping, vulnerability assessment, decision support, community engagement, and cross-border collaboration. Future research directions include integrating advanced technologies, fine-tuning spatial resolution, multisensor data fusion, social–environmental integration, climate change adaptation strategies, community-centric early warning systems, policy integration, ethics and privacy protocols, and capacity-building initiatives. This systematic review provides extensive knowledge and offers valuable insights to help researchers, policymakers, practitioners, and communities address the intricate problems of flood management in the dynamic landscapes of South Asia. Full article

Compaction leads to reduced crop yields, as the soil structure is destroyed. As soil structures respond differently to different degrees of compaction stress, early warnings for the risk of soil compaction caused by agricultural machinery need to be provided based on changes to the soil structure. In this study, we quantified the changes in the aeration porosity, aggregate mean weight diameter, structure coefficient, and cone index of different soil layers in response to compaction stress under different tyre axle weights (7.0 kN, 11.5 kN, 15.8 kN, and 20.4 kN) to analyse the risk of soil compaction in the topsoil layer (0–25 cm). The results showed that the compaction stresses that led to significant changes in soil structure in the 0–5 cm, 5–10 cm, 10–15 cm, and 15–20 cm soil layers were 130 kPa, 156 kPa, 111 kPa, and 103 kPa, respectively, and were significantly greater than the precompression stress of the soil in each layer. This finding proves that the changes in soil volume and structure caused by compaction are significant but not exactly equivalent; moreover, a threshold past which the stress caused by compaction results in soil structure failure still exists. Under 180 kPa of surface contact stress, the soil cone index, aeration porosity, aggregate mean weight diameters, and structure coefficient of the 0–5 cm and 5–10 cm soil layers gradually moved closer to the soil parameter levels of the subsoil layer before compaction. We suggest that the response of the soil structure to compaction stress proceeds along three stages, elastic deformation, plastic deformation without structure failure, and soil structure failure, within which soil structure failure stress and precompression stress are the two key threshold stresses. This study provides a more reliable theoretical basis upon which field managers can warn of soil compaction risk. Full article

The safety of power batteries in the automotive industry is of paramount importance and cannot be emphasized enough. As lithium-ion battery technology continues to evolve, the energy density of these batteries increases, thereby amplifying the potential risks linked to battery failures. This study explores pivotal safety challenges within the electric vehicle sector, with a particular focus on thermal runaway and gas emissions originating from lithium-ion batteries. We offer a non-dispersive infrared (NDIR) gas sensor designed to efficiently monitor battery emissions. Notably, carbon dioxide (CO₂) gas sensors are emphasized for their ability to enhance early-warning systems, facilitating the timely detection of potential issues and, in turn, improving the overall safety standards of electric vehicles. In this study, we introduce a novel CO₂ gas sensor based on the advanced pyroelectric single-crystal lead niobium magnesium titanate (PMNT), which exhibits exceptionally high pyroelectric properties compared to commercially available materials, such as lithium tantalate single crystals and lead zirconate titanate ceramics. The specific detection rate of PMNT single-crystal pyroelectric infrared detectors is more than four times higher than lithium tantalate single-crystal infrared detectors. The PMNT single-crystal NDIR gas detector is used to monitor thermal runaway in lithium-ion batteries, enabling the rapid and highly accurate detection of gases released by the battery. This research offers an in-depth exploration of real-time monitoring for power battery safety, utilizing the cutting-edge pyroelectric single-crystal gas sensor. Beyond providing valuable insights, the study also presents practical recommendations for mitigating the risks of thermal runaway in lithium-ion batteries, with a particular emphasis on the development of effective warning systems. Full article

Crimean–Congo hemorrhagic fever (CCHF) is a serious tick-borne disease with a wide geographical distribution. Classified as a level 4 biosecurity risk pathogen, CCHF can be transmitted cross-species due to its aerosol infectivity and ability to cause severe hemorrhagic fever outbreaks with high morbidity and mortality. However, current methods for detecting anti-CCHFV antibodies are limited. This study aimed to develop a novel luciferase immunosorbent assay (LISA) for the detection of CCHFV-specific IgG antibodies. We designed specific antigenic fragments of the nucleoprotein and evaluated their sensitivity and specificity in detecting IgG in serum samples from mice and horses. In addition, we compared the efficacy of our LISA to a commercial enzyme-linked immunosorbent assay (ELISA). Our results demonstrated that the optimal antigen for detecting anti-CCHFV IgG was located within the stalk cut-off domain of the nucleoprotein. The LISA exhibited high specificity for serum samples from indicated species and significantly higher sensitivity (at least 128 times) compared with the commercial ELISA. The proposed CCHFV-LISA has the potential to facilitate serological diagnosis and epidemiological investigation of CCHFV in natural foci, providing valuable technical support for surveillance and early warning of this disease. Full article

In recent years, implantable sensors have been extensively researched since they allow localized sensing at an area of interest (e.g., within the vicinity of a surgical site or other implant). They allow unobtrusive and potentially continuous sensing, enabling greater specificity, early warning capabilities, and thus timely clinical intervention. Wireless remote interrogation of the implanted sensor is typically achieved using radio frequency (RF), inductive coupling or ultrasound through an external device. Two categories of implantable sensors are available, namely active and passive. Active sensors offer greater capabilities, such as on-node signal and data processing, multiplexing and multimodal sensing, while also allowing lower detection limits, the possibility to encode patient sensitive information and bidirectional communication. However, they require an energy source to operate. Battery implantation, and maintenance, remains a very important constraint in many implantable applications even though energy can be provided wirelessly through the external device, in some cases. On the other hand, passive sensors offer the possibility of detection without the need for a local energy source or active electronics. They also offer significant advantages in the areas of system complexity, cost and size. In this review, implantable passive sensor technologies will be discussed along with their communication and readout schemes. Materials, detection strategies and clinical applications of passive sensors will be described. Advantages over active sensor technologies will be highlighted, as well as critical aspects related to packaging and biocompatibility. Full article

Air pollution in Jakarta, one of the most polluted cities globally, has reached critical levels, with PM_2.5 concentrations exceeding the WHO guidelines and posing significant health risks. Accurate forecasting of PM_2.5 is crucial for effective air quality management and public health interventions. PM_2.5 exhibits significant nonlinear fluctuations; thus, this study employed two machine learning approaches: self-exciting threshold autoregressive tree (SETAR-Tree) and long short-term memory (LSTM). The SETAR-Tree model integrates regime-switching capabilities with decision tree principles to capture nonlinear patterns, while LSTM models long-term dependencies in time-series data. The results showed that: (1) SETAR-Tree outperformed LSTM, achieving lower RMSE (0.1691 in-sample, 0.2159 out-sample) and MAPE (2.83% in-sample, 2.98% out-sample) compared to LSTM’s RMSE (0.2038 in-sample, 0.2399 out-sample) and MAPE (3.48% in-sample, 4.05% out-sample); (2) SETAR-Tree demonstrated better responsiveness to sudden regime changes, capturing complex pollution patterns influenced by meteorological and anthropogenic factors; (3) PM_2.5 in Jakarta often exceeds the WHO limits, highlighting this study’s importance in supporting strategic planning and providing an early warning system to reduce outdoor activity during extreme pollution. Full article

Stress events induced by global warming pose severe threats to vegetation health. Assessing the impact of these stress events on the health and growth of vegetation ecosystems in China is crucial. This study constructed three vegetation health assessment systems and selected the one that most effectively reflects vegetation health. By identifying the characteristics of stress events, and employing trend analysis, sensitivity analysis, anomaly change analysis, and modified residual analysis, this study explores the dynamic changes in vegetation health and their responses to stress events across China from 2001 to 2020. The results indicate that the Pressure–State–Response (PSR) model has the best evaluation performance, achieving the highest fit to Solar-Induced Chlorophyll Fluorescence (SIF) with an goodness of fit (R²) of up to 0.74. Overall, vegetation health exhibits more negative anomalies than positive ones and shows the strongest positive sensitivity to Cumulative Precipitation Anomaly (CPA) and the strongest negative sensitivity to Cumulative Heat (CH). Among different vegetation types, alpine vegetation has the highest stability in health, while meadows and grasslands are the most sensitive to stress events. Additionally, stress events have a greater contribution rate to vegetation health than other events. Our findings will provide important data for climate change adaptation policies and extreme environmental early warning while also contributing to the formulation of policies aimed at improving vegetation health. These results are of significant importance for enhancing carbon sequestration capacity, refining carbon market policies, and promoting the sustainable development of ecosystems. Full article

The timely and accurate monitoring of wildfires and other sudden natural disasters is crucial for safeguarding the safety of residents and their property. Satellite imagery for wildfire monitoring offers a unique opportunity to obtain near-real-time disaster information through rapid, large-scale remote sensing mapping. However, existing wildfire monitoring methods are constrained by the temporal and spatial limitations of remote sensing imagery, preventing comprehensive fulfillment of the need for high temporal and spatial resolution in wildfire monitoring and early warning. To address this gap, we propose a rapid, high-precision wildfire extraction method without the need for training—SAFE. SAFE combines the generalization capabilities of the Segmentation Anything Model (SAM) and the high temporal effectiveness of hotspot product data such as MODIS and VIIRS. SAFE employs a two-step localization strategy to incrementally identify burned areas and pixels in post-wildfire imagery, thereby reducing computational load and providing high-resolution wildfire impact areas. The high-resolution burned area data generated by SAFE can subsequently be used to train lightweight regional wildfire extraction models, establishing high-precision detection and extraction models applicable to various regions, ultimately reducing undetected areas. We validated this method in four test regions representing two typical wildfire scenarios—grassland and forest. The results showed that SAFE’s F1-score was, on average, 9.37% higher than alternative methods. Additionally, the application of SAFE in large-scale disaster scenarios demonstrated its potential capability to detect the fine spatial distribution of wildfire impacts on a global scale. Full article

The scarcity of landslide samples poses a critical challenge, impeding the broad application of machine learning techniques in landslide susceptibility assessment (LSA). To address this issue, this study introduces a novel approach leveraging a deep convolutional generative adversarial network (DCGAN) for data augmentation aimed at enhancing the efficacy of various machine learning methods in LSA, including support vector machines (SVMs), convolutional neural networks (CNNs), and residual neural networks (ResNets). Experimental results present substantial enhancements across all three models, with accuracy improved by 2.18%, 2.57%, and 5.28%, respectively. In-depth validation based on large landslide image data demonstrates the superiority of the DCGAN-ResNet, achieving a remarkable landslide prediction accuracy of 91.31%. Consequently, the generation of supplementary samples via the DCGAN is an effective strategy for enhancing the performance of machine learning models in LSA, underscoring the promise of this methodology in advancing early landslide warning systems in western Sichuan. Full article

Three-dimensional hydro-acoustic imaging is a research hot spot in the underwater acoustic signal processing field, which has a wide range of application prospects in marine environmental resource surveying, seabed topography and geomorphological mapping, and underwater early warning and monitoring. To solve the problem that the resolution of the current imaging sonar reduces rapidly with increase in distance and a scanning gap exists in side-scan sonar, we designed a down-looking 3D-imaging sonar with a linear array structure. The imaging scheme adopts a time-domain spatial beam-forming method with the Back Projection (BP) algorithm as the core, and the formation of a virtual plane array can effectively improve the along-track resolution. To cope with the interference of the carrier motion error on the imaging, we proposed a high-precision sub-wavelength motion compensation method based on a real-time acoustic calibration system. Simulation and real data experiments show that the motion compensation method can effectively eliminate the influence of motion error and make the imaging energy more focused, leading to higher-quality acoustic images. Under equal average energy, the maximum superimposed sound intensity values in the imaging results increased by 20.75 dB and 6.57 dB, respectively, for simulation and real data. After motion compensation, the resolution of this imaging system reached 3 cm × 3 cm × 2.5 cm @ Depth = 17 m, TBP = 30 s · Hz. Full article

Revealing geological environment resilience (GER) under seawater intrusion (SWI) hazards is a prerequisite for solving groundwater resource depletion, land salinization, and ecological degradation in coastal cities. This study applies the resilience design approach based on urban complex adaptive systems theory to understand the impact of SWI on the geological environment. Taking SWI as the research object, the GER evaluation method under SWI disaster was established by selecting five elastic indexes: disturbance intensity, geological environment vulnerability, stress resistance, recovery, and adaptability. This method is used to evaluate the GER level of the coastal areas of Shenzhen in recent years under the impact of SWI hazards. The study found that there is a negative correlation between the intensity of disturbance and precipitation amount. The vulnerability is greater the closer the distance to the coastline and the shallower the depth of bedrock burial. Resistance is composed of early warning ability and disaster prevention ability, and the result is 10.07, which belongs to the medium level. The recovery is 1.49, which is at a relatively high level, indicating a high capacity for restoration ability. The adaptability increased from 3.03 to 3.13, so that the area of seawater intrusion is becoming smaller. GER is affected by precipitation amount and depth of bedrock burial; the greater the precipitation and the shallower the bedrock burial, the lower the GER. Precipitation amount significantly impacts the SWI situation in the eastern coastal area of Shenzhen. In the central region, the impact of precipitation on GER is less significant. However, in the western region, the depth of bedrock burial primarily affects GER. Compared to completely weathered granite, Pleistocene fluvial plain sediments are more susceptible to SWI effects in freshwater environments. This study contributes to a deeper understanding of the impact of SWI on the geological environment in coastal areas, providing decision-makers with the necessary knowledge to develop targeted and effective governance and prevention strategies. Full article

Arterial hypertension is one of the world’s leading risk factors for death and disability. With the number of people living with this disease doubling between 1990 and 2019 from 650 million to 1.3 billion, it is a global burden that increases mortality from cardiovascular and kidney diseases. It is extremely important to use all possible diagnostic methods, indicating the possibility of early detection that subsequently leads to effective prevention of disease development. The phenomenon called hypertensive response to exercise (HRE) is one such method. The HRE is defined as a pathological, excessive increase in blood pressure as a result of exposure to the stressor, which is physical exercise. There is no consensus about precise cutoffs in the definition of this condition, which is most commonly diagnosed based on systolic blood pressure (SBP) ≥ 210 mm Hg in men and ≥190 mm Hg in women at peak exercise intensity. The fact that exercise hypotension is a pathologic sign is universally accepted. Accumulating data deliver the information that HRE is also connected to higher overall cardiovascular risk. It was demonstrated that HRE is associated with functional and structural impairment of the left ventricle and the future development of hypertension. HRE should act as a warning signal of increased cardiovascular risk, leading to the need for profound clinical care. Full article

Droughts are long-term natural disasters and encompass many unknown factors. Herein, yearly and seasonal standardized precipitation evapotranspiration index (SPEI) values were calculated by analyzing monthly temperature and precipitation data from 1971 to 2020. A cloud model was employed to obtain the spatiotemporal variations in the yearly distribution of drought weather. The cross-wavelet transform results revealed the relationship between the SPEI and atmospheric circulations. The results indicated that the average reduction rates of the SPEI-3 and SPEI-12 in Yinshanbeilu were 0.091 and 0.065 yr⁻¹, respectively, and the annual drought occurrence frequency reached 30.37%. The annual station ratio and drought intensity showed increasing trends, whereas the degree of drought slightly decreased. The overall drought conditions indicated an increasing trend, the entropy (E_n) and hyper entropy (H_e) values demonstrated increasing trends, and the expectation (E_x) showed a downward trend. The fuzziness and randomness of the drought distribution were relatively low, and the certainty of drought was relatively easy to measure. The variation in the drought distribution was relatively low. There were resonance cycles between the SPEI and various teleconnection factors. The Pacific Decadal Oscillation (PDO) and the El Niño–Southern Oscillation (ENSO) exhibited greater resonance interactions with the SPEI than did other teleconnection factors. The cloud model exhibits satisfactory application prospects in Yinshanbeilu and provides a systematic basis for early warning, prevention, and reduction in drought disasters in this region. Full article