Search Results (1,193)

Outbreaks of the Crown-of-Thorns Starfish (COTS, Acanthaster spp.) are a leading biological driver of coral cover loss, making timely and accurate population monitoring essential for reef management. Conventional diver-based surveys are labor-intensive and prone to missed detections, motivating automated detection from underwater imagery. However, COTS detection in complex underwater scenes still faces three major challenges. First, COTS individuals are often very small and carry limited discriminative information, making them inherently difficult to detect. Second, low underwater contrast and complex coral textures blur target boundaries and cause targets to be easily confused with the background. Third, ecological monitoring values recall more highly than precision—missing a COTS individual is far more costly than a false alarm—yet the recall of existing detectors remains insufficient. To address these challenges, we propose HiFi-Det (High-resolution Frequency-integration Detector), a collaborative multi-scale frequency-domain feature optimization method built on YOLO11. HiFi-Det integrates three complementary enhancements: a high-resolution detection branch that strengthens feature representation for small targets; wavelet transform convolution (WTConv) modules in the backbone and neck that apply band-separated processing in the wavelet domain to improve discrimination of COTS targets from low-contrast, textured coral backgrounds; and a WIoUv3 bounding box regression loss that dynamically focuses on ordinary-quality samples to improve recall while maintaining precision. On the public Great Barrier Reef dataset, HiFi-Det attains 81.02% F2 and 87.54% mAP@50, surpassing the YOLO11 baseline by 3.00% and 2.57%, respectively, while keeping the parameter count essentially unchanged relative to the YOLO11s baseline (within 3%), so that the accuracy gains are obtained without inflating model size. Ablation studies confirm the synergy of the three components: the high-resolution branch preserves spatial details, WTConv suppresses background textures, and WIoUv3 further curbs false positives while sustaining high recall. Applying the same recipe to a larger YOLO11m backbone yields HiFi-Det-m, which likewise improves over that backbone in both F2 and recall, indicating that the approach is a transferable recipe rather than a single fixed architecture. These results show that task-specific architectural and training designs can effectively adapt generic detectors to the demands of underwater ecological monitoring. Full article

Fine-grained classification of urban green spaces (UGSs) is important for urban ecological assessment and management but remains challenging because of spectral similarity among vegetation types and inaccurate object delineation in complex urban environments. This study proposes a pixel-to-object framework that combines semantic segmentation-guided object construction with multi-periodic phenological modeling. A semantic green-space mask derived from 0.27 m very-high-resolution imagery constrains superpixel segmentation to generate spatially coherent, boundary-aware green space object-level patches (GSOPs). Pixel-level temporal representations are then derived from Sentinel-2 normalized difference vegetation index (NDVI) time series using TimesNet, aggregated into GSOP-level phenological features, and combined with spatial attributes to classify urban trees, grasslands, and farmlands. Applied to the built-up area of Chengdu, China, the framework achieved an overall accuracy of 91.6%, with F1-scores of 92.5%, 91.9%, and 87.6% for urban trees, grasslands, and farmlands, respectively. Ablation experiments showed that removing phenological features reduced overall accuracy by 13.1 percentage points and decreased the F1-scores of grasslands and farmlands by 16.0 and 23.0 percentage points, respectively. These results demonstrate that semantically constrained object delineation and phenological information jointly reduce boundary fragmentation and improve the discrimination of spectrally similar urban vegetation types. Full article

In this study, the author investigated the wake characteristics of a shrouded wind turbine and performed a high-resolution large-eddy simulation (LES) investigation using a supercomputer. The turbulent kinetic energy (TKE) of the shrouded wind turbine was significantly greater than that of a conventional wind turbine (non-shrouded wind turbine) in the range of x/D = 0 to 5 (where x is the distance downstream of the hub center and D is the rotor diameter), due to large-scale vortices generated and released from the brim of the shrouded wind turbine and the separated flow from the nacelle. For example, the TKE was 4.4 times larger at x/D = 2. Significant differences were also observed in the wake width. The wake width of the shrouded wind turbine was approximately 2.4 times wider than that of the conventional wind turbine. In contrast, the shrouded and conventional wind turbines exhibited nearly similar behaviors in the far-wake region downstream of x/D = 5. Furthermore, a calculation was performed for the shrouded wind turbine while omitting the brim connected to the diffuser. While flow separation from the diffuser is clearly observed, the numerical results for the shrouded wind turbine without the brim connected to the diffuser showed a flow pattern very similar to that of a conventional wind turbine. Full article

The groundwater resources are increasingly stressed in the Islamabad–Rawalpindi metropolitan area of Pakistan due to unplanned urbanization, growth of industries, and inadequate waste management. In this study, the aquifer vulnerability was evaluated in the productive alluvial zone of Islamabad Watershed using a Geographic Information System (GIS)-based DRASTIC model and critically comparing it with independent measured contamination of groundwater, which is a common weakness in many machine-learning-based DRASTIC studies considering the vulnerability index as the model input. The data from 21 boreholes supplied by the Capital Development Authority (CDA) were used to map seven hydrogeological parameters in ArcGIS Pro at a 30 m resolution. The DRASTIC Index values ranged from 69 to 188, with 12.9% of the mapped watershed (209.3 km²) being rated as Very High vulnerability, mainly in the shallow western urban alluvium where water tables are below 5 m. Single-parameter sensitivity analysis showed that the most influential factors of the index were impact of the vadose zone (S_i = 1.14) and depth to water table (S_i = 1.09). A Random Forest model was trained on independently measured nitrate instead of the DRASTIC Index, but had a poor predictive skill (cross-validated R² = 0.08), and the SHapley Additive exPlanations (SHAP) analysis suggested that increased vulnerability (shallow water table and high recharge) was correlated with lower nitrate concentrations. The inverse relationship between groundwater intrinsic vulnerability and measured nitrate was statistically significant when compared to 233 groundwater samples collected at the same locations during two different campaigns (2018 and 2024) (pooled Pearson r = −0.27, p < 0.001; Spearman ρ = −0.19, p = 0.007; Kruskal–Wallis H = 14.10, p = 0.003). Levels of nitrate in both Low and Moderate vulnerability zones (6.0 and 7.7 mg/L, respectively) were higher than in Very High zones (3.4 mg/L). The inverse direction was consistent across both campaigns and robustly significant in the 2024 dataset (ρ = −0.33, p < 0.001), which covered a wider contamination gradient; in the 2018 dataset, only the parametric test was significant. Nitrate showed no significant difference between land-use classes (H = 7.23, p = 0.065) and was found as a few individual high concentrations, suggesting that these were not diffuse loading issues or intrinsic susceptibility, but were likely influenced by point sources. These results show that intrinsic DRASTIC vulnerability is useful to identify areas vulnerable to potential future contamination, but does not explain the current distribution of contamination in this aquifer, which is influenced by point-source loading and residence-time effects. To provide effective groundwater protection, intrinsic vulnerability assessment must be complemented with specific monitoring of point sources. Full article

The use of Satellite-Derived Bathymetry (SDB) constitutes an efficient, cost-effective, time-saving, and scalable approach for generating high-resolution shallow-water bathymetry. In this context, SDB can be proven to be a valuable method for supporting bathymetric surveys of submerged archaeological sites. This paper focuses on the ancient city of Asopos, located on the coast of the modern village of Plytra in Laconia, SE Peloponnese, Greece. The site occupies an extensive coastal area characterised by numerous ancient remains and geomorphological features distributed along the shoreline and within the nearshore zone. To achieve the objective of producing a very-high-resolution SDB map, one empirical and four machine learning approaches were evaluated: (a) the quadratic Stumpf band-ratio model; (b) Categorical Boosting; (c) Random Forest; (d) Extreme Gradient Boosting, with all models trained using the same three optimal spectral band ratios; and (e) the CatBoost model trained using principal components derived from a Principal Component Analysis of the multispectral dataset. These approaches were developed using multispectral WorldView-2 imagery and 208 in situ depth measurements as ground truth, collected across a different range of depths and distances from the coastline. Pan-sharpened imagery was used for visual interpretation. Model accuracy was assessed using an additional validation dataset of 57 depth measurements. This combined approach demonstrates that VHR SDB mapping of the nearshore zone is highly effective when employing three band ratios involving the blue, green, and yellow bands, achieving a root mean squared error of less than 1 m, and can therefore serve as a reliable method for shallow-water geoarchaeological investigations. Full article

Freely meandering rivers flow through gently sloping plains composed of loess and fluvial sediments. Low-gradient alluvial plains are formed without the influence of landscape features such as rock outcrops or other features that distort the flow path. The channel configurations of such rivers, and in particular the size of meanders and oxbow lakes, depend on the average long-term water discharge. Large rivers form large meanders, while small rivers form correspondingly small ones. Morphological analysis of river channel configurations can offer a metric for estimating the average long-term water discharge of a river based solely on the sinuosity of its channel. The study examined six freely meandering rivers in Kazakhstan, with discharges ranging from 4.5 to 760 m³/s and channel slopes from 0.005 to 0.06%. The morphological analysis of river channels was based on the Relative Elevation Model, specifically its version based on the Copernicus Global Digital Elevation Model, with a spatial resolution of 30 m. River channel configurations were approximated using a set of inscribed circles, the diameters of which formed the basis for the river’s average long-term water discharge metric. The largest diameter circles, which could support the river channel with a sector of at least 135°, were expertly inscribed into river bends. The diameters of the inscribed circles within these sets varied from four times for small rivers to ten times for large rivers. These sets of circles, sorted by size, can characterize the average long-term water discharge of the analyzed rivers. For example, a sample of average median values of inscribed circle diameters has a high correlation with the average long-term water discharge, with a linear approximation reliability of R² = 0.997. The scope of the developed method for assessing the average long-term water discharge of freely meandering rivers includes retrospective analysis of changes in average long-term average long-term water discharge. This can provide significant historical depth of analysis, spanning centuries and millennia, since the analysis is based on describing the results of very slow processes of natural deformation of river channels. Thus, the method proposed in this study for assessing the average long-term water discharge of freely meandering rivers based on morphological analysis of their channel configurations expands the arsenal of tools for reconstructing certain paleoclimate elements related to the hydrology of territories. Full article

High-precision flood identification from high-resolution remote sensing images using deep learning network models is challenging. Severe cloud interference, limited receptive fields, insufficient boundary refinement and spatial detail preservation, and difficulty in accurately distinguishing water bodies from ground object shadows constrain the extraction method. Therefore, this study proposes an automatic flood information extraction method that integrates an improved Fully Convolutional Network classification and recognition model (FCN-RAM) with a rough tolerance set. First, a tolerance rough set algorithm was employed for sample data preprocessing. Subsequently, a Residual Attention Module (RAM) was introduced to optimize the U-Net architecture, dynamically adjusting the response intensity of deep features in both the channel and spatial dimensions to construct a deep learning-based FCN-RAM. Finally, comparative analyses were conducted on three high-resolution remote sensing datasets with different resolutions: Global surface water detection in Large-size very-High-resolution satellite imagery (GLH-Water), Gaofen Image Dataset (GID), and Earth Surface Water Dataset (ESWD). The results demonstrated that FCN-RAM consistently and substantially outperformed the baseline U-Net across all three datasets, achieving F1-score improvements of 10.64% (GLH-Water), 9.71% (GID), and 10.64% (ESWD), with corresponding overall accuracy gains of 9.97%, 11.15%, and 10.22%, respectively. Notably, the Intersection-over-Union (IoU) scores were elevated by 17.59% (GLH-Water), 15.66% (GID), and 13.63% (ESWD). The method also surpassed state-of-the-art models including ResNet and Water-SCNet, attaining peak overall accuracies of 98.61% (GLH-Water) and 97.37% (GID). Notably, while the proposed framework exhibits remarkable generalization across the evaluated multi-resolution benchmarks, its current validation is primarily confined to static water body delineation tasks. The model’s transferability to highly heterogeneous geographical regions with scarce training samples, as well as its extendability toward dynamic time-series flood evolution modeling, warrants further systematic investigation. The proposed method significantly improves the accuracy of waterbody information extraction, meets the requirements for high-precision information extraction from high-resolution imagery, and provides technical support for intelligent flood information extraction using high-resolution remote sensing. Full article

Very-high-resolution (VHR) multispectral satellite imagery contains rich semantic information, yet its real-time transmission is constrained by limited satellite-to-ground bandwidth and dynamic channel impairments. Conventional communication schemes prioritize pixel-level reconstruction, resulting in large transmission overhead and poor robustness under unfavorable channel conditions. To address these challenges, an end-to-end task-oriented semantic communication framework for remote sensing downstream recognition tasks, termed Semantic Transmission Architecture for Remote Sensing (STARS), is proposed. To improve transmission efficiency for very-high-resolution remote sensing images with highly redundant background regions, a Semantic Feature Reweighting Module (SFRM) is introduced to dynamically evaluate token-level semantic importance and adaptively allocate transmission resources to task-critical features. Furthermore, vector quantization and a practical digital transmission chain are jointly integrated to achieve efficient semantic compression, while dynamic channel variations are incorporated during training to improve robustness under fading channel conditions. Experimental results on the DOTA dataset demonstrate that STARS consistently outperforms conventional schemes and existing semantic baselines under Rician fading channels, validating the effectiveness of semantic-aware feature allocation for bandwidth-efficient VHR imagery transmission. Full article

Artificial intelligence (AI) has become an integral part of our daily lives as a most extraordinary assistant for providing information and carrying out various tasks. We can have an extensive amount of work done by a dedicated AI that we cannot do ourselves due to our easily distracted minds. AI is given vast amounts of information that we have been gathering for a long time. Using that information, it can analyze large amounts of data and sort it into particular topics and narrow down the solutions or choices we are interested in. With highly sophisticated high-resolution ground- and space-based telescopes and observatories, we have been gathering huge amounts of data. The analysis of these data will require both a considerable amount of manpower and time. Both of these factors can be reduced easily if we train dedicated AIs for spectral line identification and plasma modeling for analysis. While high-accuracy atomic data are needed for precise astrophysical plasma modeling, there are many gaps in the data due to computations of these high-accuracy data being complex and numerically challenging and requiring a significant amount of time. These problems can also be tackled to a very good approximation by training AI for simulations of existing high-accuracy data and producing data that are missing. A general algorithm for such predicted data and its implementation in astrophysical applications is presented. A recent application of AI in measuring oxygen abundance in galaxies is also presented. Full article

The accelerated pace of urbanization across the Global South calls for precise, real-time building footprint data to underpin effective urban governance and enhance disaster resilience. Conventional mapping approaches, however, suffer from inefficiency in data acquisition and updating. Although Volunteered Geographic Information (VGI) provides a crowdsourced solution for geospatial data collection, it is commonly hindered by significant heterogeneity—manifested in inconsistent data completeness, positional inaccuracies and poor topological consistency across different datasets. To address these critical limitations, this study proposes an intelligent geospatial agent framework designed to autonomously fuse building data from multiple heterogeneous sources, including VGI, Very High-Resolution (VHR) satellite imagery, and Light Detection and Ranging (LiDAR) data. This study’s core innovative points are embodied in three key modules: a supervised VGI quality verification module that leverages the Random Forest model to evaluate the reliability of individual building feature elements; a hybrid building extraction engine which integrates LiDAR data with the Segment Anything Model (SAM) to realize zero-shot building extraction; and a cognitive rule engine that adopts Multi-Criteria Decision Analysis (MCDA) for the intelligent resolution of spatial conflicts. Comprehensive validation experiments were conducted in two African cities experiencing rapid urbanization—Kigali and Dar es Salaam. The results show that the proposed framework boosts data completeness by more than 29% and attains a fused dataset F1-Score of 0.919, effectively converting incomplete VGI data into a geospatial resource with near-official authoritative quality. Full article

Background and Objectives: Electronic glycemic management systems (eGMSs) are increasingly used to guide intravenous insulin infusion for hospitalized patients with diabetic ketoacidosis (DKA), but available comparative evidence remains non-randomized and clinically heterogeneous and includes multiple algorithmically distinct platforms. Methods and Materials: We conducted a systematic review and meta-analysis in accordance with PRISMA guidance and a prospectively registered protocol (PROSPERO: 2025 CRD420251019614). Seven non-randomized studies comprising 3874 hospitalized patients were included; six studies contributed data to the primary meta-analysis of time to DKA resolution. The primary outcome was time to DKA resolution. Secondary outcomes included ICU length of stay (LOS), hospital LOS, duration of insulin infusion, and hypoglycemia (mild and severe). Random-effects models were applied. Results: Six studies contributed to the primary meta-analysis of time to DKA resolution; across the review, seven studies included 3874 patients. The pooled analysis showed no statistically significant difference between groups (SMD −0.04, 95% CI −0.22 to 0.14; I² = 74.8%). Recorded hypoglycemia was lower among patients managed with eGMS; however, these estimates should be interpreted cautiously because of heterogeneity, serious-to-critical risk of bias, and potential differences in glucose-monitoring and documentation practices. ICU and hospital LOSs and duration of insulin infusion showed no statistically significant differences overall, with heterogeneity across studies. Conclusions: In available non-randomized evidence, eGMS-guided insulin infusion was not associated with a clear difference in time to DKA resolution, ICU length of stay, hospital length of stay, or insulin infusion duration compared with conventional protocols. Lower recorded hypoglycemia was observed with eGMSs, but this finding should be considered hypothesis-generating because of serious-to-critical risk of bias, residual confounding, substantial heterogeneity, and possible detection bias. High-quality randomized trials or target trial emulation studies are needed before recommending widespread adoption, as certainty of evidence was very low across all outcomes. Full article

With the increasing availability of high-resolution (<50 m) spaceborne night time light imagery, it is now becoming more feasible to examine the correspondence between spaceborne and ground-based measurements of night lights. However, so far there have been very few studies that have conducted a ground-based campaign of night time brightness measurements during the overpass of a night light-sensitive satellite. Here we tested whether the correspondence between measurements is higher when ground-based measurements are conducted at the same time as the satellite overpass. We conducted measurements using a LANcube photometer along the same route on two consecutive nights (27–28 August 2025) in Brisbane, Australia, and compared them with an SDGSAT-1 (10–40 m) and Haishao-1 (10 m) images acquired concurrently in the evening and with an early morning ISS photo (8 m) acquired three months earlier. We found the correlation between ground-based and spaceborne measurements was not higher for simultaneous measurements, and the explanatory power of our model predicting night time brightness as measured from space increased when including horizontal and upwards ground-based brightness measurements alongside variables of canopy height, land use and road hierarchy. We confirmed the importance of multidirectional ground measurements and urban structure for understanding night time brightness levels measured from space. Full article

Background: Recent technology improvements have enabled desorption electrospray ionisation (DESI) mass spectrometry imaging to achieve down to 5 µm (pixel) image resolution. However, operating at this resolution introduces challenges, particularly regarding increased total analysis time and the need for sufficient instrument sensitivity to detect analytes from very small tissue areas. Methods: High mass and image resolution DESI imaging was performed on rat brain tissue using a Xevo™ MRT benchtop mass spectrometer equipped with a multi-reflecting time-of-flight mass analyser and a DESI XS source. Data acquisition was conducted at speeds of up to 100 Hz. Sensitivity was assessed using a dilution series of five Active Pharmaceutical Ingredients (APIs) spotted onto porcine liver tissue. Signal detection limits were evaluated using extracted ion chromatograms (XICs) with signal-to-noise (S/N) calculations against blank samples. Additionally, enhanced duty cycle (EDC) was applied to evaluate improvements in analyte signal intensity across specific mass ranges in both positive and negative ionisation modes. Results: At acquisition speeds of up to 100 Hz, excellent data quality was achieved, with signal intensity remaining suitable for analytical applications. All five tested APIs were detectable at concentrations of 25 pg/mm². Three of the five compounds were further detected at concentrations as low as 2.5 pg/mm², with signal-to-noise ratios greater than 5. The application of EDC resulted in a significant increase in analyte signal intensity within the targeted mass ranges, particularly for small molecule endogenous metabolites and lipids, in both ionisation modes. Furthermore, the system demonstrated substantially improved spectral quality, achieving mass resolution up to 100,000 FWHM. This enabled the resolution of previously indistinguishable analytes with significantly improved mass accuracy compared to systems operating at approximately 30,000 FWHM. Conclusions: The Xevo™ MRT mass spectrometer with DESI XS source enables high-resolution DESI imaging at speeds up to 100 Hz without compromising data quality or sensitivity. The system demonstrates excellent detection limits for pharmaceutical compounds and improved performance through enhanced duty cycle operation. Overall, the combination of high spatial resolution, increased mass resolution, and improved spectral quality allows for more accurate analyte differentiation, representing a significant advancement over lower-resolution systems. Full article

Monitoring urban biodiversity is essential for designing resilient and sustainable cities. Urban trees provide a wide range of ecosystem services (ESs), including air pollution reduction, urban heat island mitigation, and psychological benefits for citizens. Accurate and updated tree inventories are therefore essential tools for urban environmental monitoring. However, existing urban tree inventories are often incomplete or outdated, especially in private areas, limiting accurate ES assessment and urban planning. Earth observation satellite missions, particularly very-high-resolution multispectral (VHR-MS) imagery, offer a valuable alternative to field surveys for gathering information on urban environments. This work proposes a deep learning (DL) framework based on VHR-MS satellite imagery for the automatic generation of accurate urban tree inventories. DL models reduce human effort and save operational time by automatically learning complex representations and patterns from satellite imagery. The proposed encoder–decoder architecture extends prior point-based detection approaches by integrating a ResNet-50 backbone and a percentile-based threshold calibration procedure. Given the lack of suitable training data covering heterogeneous and densely vegetated urban environments, a dedicated dataset was constructed from VHR-MS satellite imagery acquired over the Lombardy region (Italy). The dataset encompasses a wide range of land uses and land covers, including residential and industrial zones, public parks, private gardens, and agricultural areas. Through the photointerpretation of more than 2800 images, precise coordinates for more than 50,000 manually annotated trees were obtained. The DL model is trained with point-level annotations, enabling precise localization of individual trees while reducing annotation ambiguity in dense urban contexts. On the Lombardy dataset at 30 cm/px resolution, the proposed framework achieves 86.72% Precision, 66.92% Recall, an F1-score of 75.54%, and a localization error of 1.473 m. Full article

(1) Background: For the study of highly magnetized neutron stars observed as magnetars and to quantify the effect of this intense magnetic field on the star’s structure and shape, which can be particularly relevant for the study of the emission of continuous gravitational waves, both numerical and perturbative approaches have been developed. (2) Methods: We compare these two approaches in General Relativity with the limitation to the case where the magnetic field has a purely poloidal structure. The perturbative one assumes that the deformation induced by the magnetic field is small and that this field arises only from dipole currents. The fully numerical one is based on the lorene library. (3) Results: We used both approaches to compute the magnetic-field distribution and the deformation of the star, varying the value of the magnetic field at the pole, the compactness of the star and its equation of state. (4) Conclusions: Whereas the perturbative approach breaks down for very high polar magnetic-field values (typically above a few times ${10}^{16}$ G), it achieves very good results for observed values, even in magnetars. On the contrary, the numerical code exhibits resolution problems for relatively low magnetic-field values (typically ${10}^{10}$ G), which translates into imprecise computation of the star’s deformation and mass quadrupole moment. Full article