Search Results (7,703)

Complex-valued convolutional neural networks (CVCNNs) have demonstrated strong capabilities for polarimetric synthetic aperture radar (PolSAR) image classification by effectively integrating both amplitude and phase information inherent in polarimetric data. However, their practical deployment faces significant challenges due to high computational costs and performance degradation caused by extremely limited labeled samples. To address these challenges, a lightweight CV Siamese network (LCVSNet) is proposed for few-shot PolSAR image classification. Considering the constraints of limited hardware resources in practical applications, simple one-dimensional (1D) CV convolutions along the scattering dimension are combined with two-dimensional (2D) lightweight CV convolutions. In this way, the inter-element dependencies of polarimetric coherency matrix and the spatial correlations between neighboring units can be captured effectively, while simultaneously reducing computational costs. Furthermore, LCVSNet incorporates a contrastive learning (CL) projection head to explicitly optimize the feature space. This optimization can effectively enhance the feature discriminability, leading to accurate classification with a limited number of labeled samples. Experiments on three real PolSAR datasets demonstrate the effectiveness and practical utility of LCVSNet for PolSAR image classification with a small number of labeled samples. Full article

This review synthesizes advances in nondestructive DNA extraction from teeth, emphasizing their importance in forensics and archaeogenetics. Because of their mineralized structure and resistance to diagenesis, teeth remain vital for human identification when other tissues are unavailable or degraded. Modern protocols targeting dental cementum have shown high success rates in retrieving nuclear DNA while maintaining specimen integrity, supporting ethical standards, and enabling additional morphological and isotopic analyses. Nondestructive extraction methods produce DNA yields comparable to—or in some archaeological cases, greater than—those of traditional destructive approaches, while ensuring strict contamination control and minimal physical impact. Cementum is a reliable source of DNA in aged and degraded teeth, although the petrous part of the temporal bone still represents the best option under extreme preservation conditions. These results highlight the need for context-specific sampling strategies that balance analytical goals with the preservation of museum collections. Future efforts include testing nondestructive protocols across various forensic scenarios and creating predictive models for DNA preservation. Overall, these developments promote ethical, effective, and sustainable practices in human genomic analysis. Full article

Brain magnetic resonance imaging (MRI) is highly susceptible to motion artifacts that degrade fine structural details and undermine quantitative analysis. Conventional U-Net-based deep learning approaches for motion artifact reduction typically operate only in the image domain and are often trained on data with simplified motion patterns, thereby limiting physical plausibility and generalization. We propose Sim-DDNet, a simulation-data-based dual-domain network that combines k-space-based motion simulation with a joint image-k-space reconstruction architecture. Motion-corrupted data were generated from T2-weighted Alzheimer’s Disease Neuroimaging Initiative brain MR scans using a k-space replacement scheme with three to five random rotational and translational events per volume, yielding 69,283 paired samples (49,852/6969/12,462 for training/validation/testing). Sim-DDNet integrates a real-valued U-Net-like image branch and a complex-valued k-space branch using cross attention, FiLM-based feature modulation, soft data consistency, and composite loss comprising L1, structural similarity index measure (SSIM), perceptual, and k-space-weighted terms. On the independent test set, Sim-DDNet achieved a peak signal-to-noise ratio of 31.05 dB, SSIM of 0.85, and gradient magnitude similarity deviation of 0.077, consistently outperforming U-Net and U-Net++ across all three metrics while producing less blurring, fewer residual ghost/streak artifacts, and reduced hallucination of non-existent structures. These results indicate that dual-domain, data-consistency-aware learning, which explicitly exploits k-space information, is a promising approach for physically plausible motion artifact correction in brain MRI. Full article

Unsupervised anomaly detection has been extensively studied. However, most existing methods are designed for either numerical or nominal data, which struggle to detect anomalies effectively in real-world mixed-type datasets. Fuzzy information granulation is a key concept in granular computing, which offers a potent framework for managing uncertainty in mixed-type data and provides a viable pathway for unsupervised anomaly detection. Nevertheless, conventional fuzzy information granulation-based detection methods often model only simple, linear fuzzy relations between samples. This limitation prevents them from capturing the complex, nonlinear structures inherent in the data, leading to a degradation in detection performance. To address these shortcomings, we propose a Hybrid Kernelized Fuzzy Density-based anomaly detector (HKFD). HKFD pioneers a hybrid kernelized fuzzy relation by integrating a hybrid distance metric with kernel methods. This new relation allows us to define a hybrid kernelized fuzzy density for each sample within every feature subspace, effectively capturing the local data dispersion. Crucially, we introduce an information-theoretic weighting mechanism. By calculating the fuzzy information entropy of each feature’s distribution, HKFD automatically assigns higher weights to more informative feature subspaces that contribute more to identifying anomalies. The final anomaly factor is then calculated by the weighted fusion of these densities. Comprehensive experiments on 20 datasets demonstrate that HKFD significantly outperforms state-of-the-art methods, achieving superior anomaly detection performance. Full article

Stabilizing thermoplastic polymers against thermal degradation is an important aspect that must be addressed during material development and becomes critical in the case of bio-polymers, which often reveal reduced thermal stability and a narrow processing temperature window. Herein, we propose a new methodology to analyze and compare the thermal stability of thermoplastic materials, exampled by several types of bio-polyesters, such as aliphatic PBS and PBSA, aliphatic-aromatic PBAT and PBST, and amorphous PHBV, and evaluate the impact of thermal stabilizer on their processability and thermal stability. The proposed method relies on multi-step torque rheometry experiments that involve controlled cycling of the tested material under varied thermal conditions, shear forces, and processing times to acquire and evaluate the changes in flow behavior of the sample after its processing. By monitoring polymer melt behavior and comparing the changes before and after repetitive processing steps, we can gain valuable insights into the material performance and stabilizing efficiency of additives. The thermal stability of polymers and the efficiency of thermal stabilizers can be assessed by means of the relative change in temperature-normalized torque, $∆\tau \left(\%\right)$, measured after different processing steps. Significantly, we demonstrate that the obtained $∆\tau \left(\%\right)$ values correlate with changes in the molar mass of neat polymers as a result of their processing. The proposed approach enables a semi-quantitative evaluation of the thermal stability of various polymers and the study of the efficiency of thermal stabilizers and their performance, providing a robust strategy for optimizing compound formulations, particularly regarding the optimal fractions required. Full article

Extreme weather conditions, specifically typhoons and strong gusts, create a highly transient environment for wind power data collection, leading to performance degradation that significantly impacts the safety and stability of the wind power system. To accurately predict wind power trends under these conditions, this paper proposes a prediction model integrating Singular Spectrum Analysis (SSA), Temporal Convolutional Network (TCN), Convolutional Neural Network (CNN), and a global average pooling layer, combined with inverse adaptive transfer learning. First, SSA is applied to reduce noise in the collected wind power operation data and extract key information. Subsequently, a prediction model is constructed based on TCN, CNN, and global average pooling. The model employs dilated causal convolutions to capture long-term dependencies and uses two-dimensional convolution kernels to extract local mutation features. Furthermore, a domain-adaptive transfer learning module is designed to adjust the model’s parameter weights via backward optimization based on the Maximum Mean Discrepancy (MMD) between the source and target domains. Experimental validation is conducted using real-world wind power operation data from a wind farm in Guangxi, containing 3000 samples sampled at 10 min intervals specifically during severe typhoon periods. Experimental results demonstrate that even with only 60% of the target data, the proposed method outperforms the traditional TCN neural network, reducing the Root Mean Square Error (RMSE) by 58.1% and improving the Coefficient of Determination (R²) by 32.7%, thereby verifying its effectiveness in data-scarce extreme scenarios. Full article

Under sparse aperture (SA) conditions, inverse synthetic aperture radar (ISAR) imaging becomes a severely ill-posed inverse problem due to undersampled and noisy measurements, leading to pronounced degradation in azimuth resolution and image quality. Although deep learning approaches have demonstrated promising performance for SA-ISAR imaging, their practical deployment is often hindered by black-box behavior, fixed network depth, high computational cost, and limited robustness under extreme operating conditions. To address these challenges, this paper proposes an ADMM Denoising Deep Equilibrium Framework (ADnDEQ) for SA-ISAR imaging. The proposed method reformulates an ADMM-based unfolding process as an implicit deep equilibrium (DEQ) model, where ADMM provides an interpretable optimization structure and a lightweight DnCNN is embedded as a learned proximal operator to enhance robustness against noise and sparse sampling. By representing the reconstruction process as the equilibrium solution of a single-layer network with shared parameters, ADnDEQ decouples forward and backward propagation, achieves constant memory complexity, and enables flexible control of inference iterations. Experimental results demonstrate that the proposed ADnDEQ framework achieves superior reconstruction quality and robustness compared with conventional layer-stacked networks, particularly under low sampling ratios and low-SNR conditions, while maintaining significantly reduced computational cost. Full article

This study investigates the creep behavior and acoustic emission (AE) characteristics of bedded coal samples under acidic water environments. Uniaxial graded creep tests coupled with AE monitoring were conducted on samples with bedding angles of 0°, 30°, 60°, and 90°, respectively. The anisotropic mechanical behavior and acoustic emission characteristics in terms of stress–strain, deformation, AE count, AE energy, and spectrum characteristics were revealed. The experimental results show that the strength of the coal samples gradually decreases as the saturation duration increases. At the same axial stress level, the axial deformation of the coal samples becomes larger with increasing saturation duration. The mechanical strength exhibits a distinct “U-shaped” relationship with the bedding angle, initially decreasing and then increasing. Correspondingly, axial deformation at a given stress level first increases and then decreases as the bedding angle increases. AE activity, particularly the AE ring count and energy, peaks at specimen failure, indicating significant fracture development. Spectral analysis revealed that under conditions of severe strength degradation (e.g., 0° bedding after 60-day saturation or 60° bedding after 30-day saturation), high-frequency, high-amplitude AE signals were absent. This suggests a shift in the dominant fracture mechanism from small-scale cracking to larger-scale fracture propagation in weakened samples. These findings offer valuable theoretical insights for the prevention and early warning of coal mine disasters. Full article

Over the last century, anthropogenic activities have contributed to habitat degradation and fragmentation but have also affected the individual health of animals. In this study, we investigated the effect of environmental differences on the gut microbiome of Herring Gulls (Larus argentatus) by collecting fresh faecal samples from ten geographically different populations in the UK, including captive and wild birds, and comparing the resulting gut microbiome diversity and composition. A significantly higher alpha diversity was identified in captive gulls than in urban and suburban gulls for the 46 sequenced samples. When comparing gut microbiome composition, urban inhabitants exhibited a higher abundance of Ligilactobacillus and a lower abundance of Streptococcus than suburban gulls. Such differences could suggest a highly polluted environment for urban-dwelling gulls, while suburban populations could have a wider foraging range and a more diverse diet. In addition, samples from Bristol, West Kirby, Gloucester and Liverpool were all characterised by a significantly higher abundance of one or more of the other bacterial taxa. The high proportion of Mycoplasma could indicate avian mycoplasmosis in the Liverpool population. This study sheds light on the understudied subject of the wild avian gut microbiome and its possible application to wildlife health and disease management. Full article

Water from mountain regions is a crucial natural resource because of its major economic, social, and environmental significance. Wildfires may disrupt the normal functioning of the hydrological cycle, limiting water resources for nearby areas and degrading water quality in mountainous regions as contaminants enter water systems from the burning of vegetation and soil. In August 2022, the Serra da Estrela mountain, situated in the Mediterranean biogeographical region, was affected by a large wildfire that consumed 270 km² of the Serra da Estrela Natural Park, often resulting in severe vegetation burn, although the soil burn severity was low to moderate in most of the area. The research objective is to assess the impact of this wildfire on the hydrogeochemistry of groundwater and surface water in the Manteigas-Covão da Ametade sector of Serra da Estrela in the context of a wildfire with limited soil burn severity. Groundwater and surface water samples were collected from October 2022 to September 2023 and were analyzed for pH, Total Organic Carbon, electrical conductivity, major ions, potentially toxic elements, iron (Fe), and Polycyclic Aromatic Hydrocarbons. A stormy event in mid-September 2022, occurring before the first sampling campaign, removed most of the ash layer and likely caused transient hydrogeochemical changes in streams. However, the analytical results from the sampled waters revealed that the post-wildfire hydrogeochemical effects are not evident. In fact, the hydrogeochemical changes observed in groundwater and surface water appear to be primarily influenced by the regular hydrological behaviour of aquifers and streams. The low to moderate soil burn severity, the high soil hydrophobicity, and the temporal distribution of precipitation explain why the hydrogeochemistry was primarily influenced by groundwater flow paths, the types and weathering of local lithologies, soil types, dilution effects following wet periods, and seasonal changes in the tributaries feeding into streams, rather than by post-wildfire effects. These outcomes provide valuable insights for water resource management and for developing strategies to mitigate wildfire impacts in mountainous environments. Full article

Glass-insulated terminals (GITs) are widely used in high-reliability microelectronic systems, where glass fall-offs in the sealing region may seriously degrade the reliability of the microelectronic component and further degrade the device reliability. Automatic inspection of such defects is challenging due to strong light reflection, irregular defect appearances, and limited defective samples. To address these issues, a coarse-to-fine machine-learning framework is proposed for glass fall-off detection in GIT images. By exploiting the circular-ring geometric prior of GITs, an adaptive sector partition scheme is introduced to divide the region of interest into sectors. Four categories of sector features, including color statistics, gray-level variations, reflective properties, and gradient distributions, are designed for coarse classification using a gradient boosting decision tree (GBDT). Furthermore, a sector neighbor (SN) feature vector is constructed from adjacent sectors to enhance fine classification. Experiments on real industrial GIT images show that the proposed method outperforms several representative inspection approaches, achieving an average IoU of 96.85%, an F1-score of 0.984, a pixel-level false alarm rate of 0.55%, and a pixel-level missed alarm rate of 35.62% at a practical inspection speed of 32.18 s per image. Full article

Welding repair of cracks in FSX-414 cobalt-based alloy, used in high-temperature components, poses significant challenges due to the presence of surface oxide films within the cracks. By comparing the formation of oxide films on the external surface and inside the narrow crevice of FSX-414 alloys preserved at 900 °C for up to 1000 h, we found that the oxide film growth rate on the external surface was slightly larger than that inside the narrow crevice, and the latter slowed down after 672 h. Additionally, the oxide films on both surfaces were mainly composed of O and Cr elements, providing excellent protection to the underlying metal and resulting in minimal internal oxidation. A compositional transition region formed between the oxide film and the base metal. The width of the transition region decreased with heating duration and was narrower in the external surface sample, leading to a steeper composition gradient between the oxide film and the inner metal. With prolonged exposure, increasing numbers of “pores” rich in W and O appeared near the oxide films, creating channels that connect the oxide layer with the internal metal and accelerate material degradation. “Pores” extended deeper into the metal within the narrow crevice compared to those on the surface. Prior to welding repair, channels composed of W and O near the oxide films must be cleaned along with the oxide layer itself, and the removal of oxide from narrow cracks poses greater difficulty. Full article

This study aimed to evaluate in situ ruminal nutrient degradation, fractional disappearance rate, and effective ruminal degradation of hydroponically sprouted barley, wheat, and triticale. Two ruminally canulated lactating cows were used in a complete randomized block design with four treatments and nine incubation times (0, 2, 4, 8, 12, 24, 48, 72, and 240 h). Treatments were corn silage (control), and sprouted barley, triticale, and wheat. Quadruplicate samples (5 g each) were placed in Dacron bags and incubated in the rumen. Then, bags were rinsed and spun, dried (48 h × 55 °C; 3 h × 105 °C), and weighed to determine residual dry matter (DM). Data were analyzed using mixed models (MIXED, SAS 9.4) with treatment, time, and their interaction as fixed effects, and cow and replicate (cow) as random effects. Denominator degrees of freedom were adjusted using the Kenward–Roger method, and means were separated by Tukey–Kramer. Significance was declared at p ≤ 0.05 and tendencies at 0.05 < p ≤ 0.10. Sprouted triticale and wheat treatments had a greater rapidly soluble fraction for DM (p < 0.01), the greatest fractional disappearance rate for DM (p < 0.01) and neutral detergent fiber (NDF; p < 0.01), and greater effective ruminal degradability (ERD) for DM (p < 0.01) and crude protein (CP; p < 0.01). Sprouted wheat also had the greatest ERD for NDF (p < 0.01). In contrast, sprouted barley had the lowest rapidly soluble fractions for DM (p < 0.01), NDF (p < 0.01), and CP (p < 0.01), lower fractional disappearance rate for DM (p < 0.01) and NDF (p < 0.01) than sprouted triticale and wheat, and the lowest ERD for DM (p < 0.01) and CP (p < 0.01). Overall, sprouted triticale and wheat had greater in situ ruminal nutrient degradation, effective ruminal degradation, and nutrient degradation kinetics, indicating their potential for inclusion in dairy cattle diets to improve nutrient degradability. Full article

Satellite communication systems experience significant signal degradation during rain events, a phenomenon that can be leveraged for meteorological applications. This study introduces a novel hybrid machine learning framework combining unsupervised clustering with cluster-specific supervised deep learning models to transform satellite signal attenuation into a predictive tool for rainfall prediction. Unlike conventional single-model approaches treating all atmospheric conditions uniformly, our methodology employs K-Means Clustering with the Elbow Method to identify four distinct atmospheric regimes based on Signal-to-Noise Ratio (SNR) patterns from a 12-m Ku-band satellite ground station at King Mongkut’s Institute of Technology Ladkrabang (KMITL), Bangkok, Thailand, combined with absolute pressure and hourly rainfall measurements. The dataset comprises 98,483 observations collected with 30-s temporal resolutions, providing comprehensive coverage of diverse tropical atmospheric conditions. The experimental platform integrates three subsystems: a receiver chain featuring a Low-Noise Block (LNB) converter and Software-Defined Radio (SDR) platform for real-time data acquisition; a control system with two-axis motorized pointing incorporating dual-encoder feedback; and a preprocessing workflow implementing data cleaning, K-Means Clustering (k = 4), Synthetic Minority Over-Sampling Technique (SMOTE) for balanced representation, and standardization. Specialized Long Short-Term Memory (LSTM) networks trained for each identified cluster enable capture of regime-specific temporal dynamics. Experimental validation demonstrates substantial performance improvements, with cluster-specific LSTM models achieving R² values exceeding 0.92 across all atmospheric regimes. Comparative analysis confirms LSTM superiority over RNN and GRU. Classification performance evaluation reveals exceptional detection capabilities with Probability of Detection ranging from 0.75 to 0.99 and False Alarm Ratios below 0.23. This work presents a scalable approach to weather radar systems for tropical regions with limited ground-based infrastructure, particularly during rapid meteorological transitions characteristic of tropical climates. Full article

Transparent polymeric materials such as poly(methyl methacrylate) (PMMA), polycarbonate (PC), and polyethylene terephthalate (PET) are widely used as glass alternatives in algal bioreactors, where optical clarity and mechanical stability are crucial. However, their long-term use is limited by surface degradation processes. Photodegradation, hydrolysis, and biofilm accumulation can reduce light transmission in the 400–700 nm range essential for photosynthesis. This study examined the aging of PMMA, PC, and PET under bioreactor conditions. Samples were exposed for 70 days to illumination, culture medium, and aquatic environments. Changes in their optical transmittance, surface roughness, and wettability were analyzed. All polymers exhibited measurable surface degradation, characterized by an average 15% loss in transparency, significant increases in surface roughness, and reduced contact angles. PMMA demonstrated the highest optical stability, maintaining strong transmission in key blue and red spectral regions, while PET performed the worst, showing low initial clarity and the steepest decline. The most severe surface degradation occurred in areas exposed to the receding liquid interface, highlighting the need for targeted cleaning and/or a reduction in the size of the liquid–vapor transition zone. Overall, the results identify PMMA and recycled PMMA (PMMA_R) as durable, cost-effective materials for transparent bioreactor walls. Full article