Search Results (1,971)

Digital image quality is crucial for reliable analysis in applications such as medical imaging, satellite remote sensing, and video surveillance. However, traditional denoising methods struggle to balance noise removal with detail preservation and lack adaptability to various types of noise. We propose a novel three-module architecture for image denoising, comprising a generator, a dual-path-UNet-based denoiser, and a discriminator. The generator creates synthetic noise patterns to augment training data, while the dual-path-UNet denoiser uses multiple receptive field modules to preserve fine details and dense feature fusion to maintain global structural integrity. The discriminator provides adversarial feedback to enhance denoising performance. This dual-path adversarial training mechanism addresses the limitations of traditional methods by simultaneously capturing both local details and global structures. Experiments on the SIDD, DND, and PolyU datasets demonstrate superior performance. We compare our architecture with the latest state-of-the-art GAN variants through comprehensive qualitative and quantitative evaluations. These results confirm the effectiveness of noise removal with minimal loss of critical image details. The proposed architecture enhances image denoising capabilities in complex noise scenarios, providing a robust solution for applications that require high image fidelity. By enhancing adaptability to various types of noise while maintaining structural integrity, this method provides a versatile tool for image processing tasks that require preserving detail. Full article

Pupillometry is commonly used to evaluate cognitive effort, attention, and facial expression response, offering valuable insights into human performance. The combination of eye tracking and facial expression data under the iMotions platform provides great opportunities for multimodal research. However, there is a lack of standardized pipelines for managing pupillometry data on a multimodal platform. Preprocessing pupil data in multimodal platforms poses challenges like timestamp misalignment, missing data, and inconsistencies across multiple data sources. To address these challenges, the authors introduced a systematic preprocessing pipeline for pupil diameter measurements collected using iMotions 10 (version 10.1.38911.4) during an endoscopy simulation task. The pipeline involves artifact removal, outlier detection using advanced methods such as the Median Absolute Deviation (MAD) and Moving Average (MA) algorithm filtering, interpolation of missing data using the Piecewise Cubic Hermite Interpolating Polynomial (PCHIP), and mean pupil diameter calculation through linear regression, as well as normalization of mean pupil diameter and integration of the pupil diameter dataset with facial expression data. By following these steps, the pipeline enhances data quality, reduces noise, and facilitates the seamless integration of pupillometry other multimodal datasets. In conclusion, this pipeline provides a detailed and organized preprocessing method that improves data reliability while preserving important information for further analysis. Full article

Injurious birds pose a significant threat to food production and the agricultural economy. To address the challenges posed by their small size, irregular shape, and frequent occlusion in complex farmland environments, this paper proposes YOLO-AWK, an improved bird detection model based on YOLOv11n. Firstly, to improve the ability of the enhanced model to recognize bird targets in complex backgrounds, we introduce the in-scale feature interaction (AIFI) module to replace the original SPPF module. Secondly, to more accurately localize and identify bird targets of different shapes and sizes, we use WIoUv3 as a new loss function. Thirdly, to remove the noise interference and improve the extraction of bird residual features, we introduce the Kolmogorov–Arnold network (KAN) module. Finally, to improve the model’s detection accuracy for small bird targets, we add a small target detection head. The experimental results show that the detection performance of YOLO-AWK on the farmland bird dataset is significantly improved, and the final precision, recall, mAP@0.5, and mAP@0.5:0.95 reach 93.9%, 91.2%, 95.8%, and 75.3%, respectively, which outperforms the original model by 2.7, 2.3, 1.6, and 3.0 percentage points, respectively. These results demonstrate that the proposed method offers a reliable and efficient technical solution for farmland injurious bird monitoring. Full article

Objective: This study aims to critically review and synthesize the existing literature on the use of voice analysis in assessing nasal obstruction, with a particular focus on acoustic parameters. Data sources: PubMed, Scopus, Web of Science, Ovid Medline, and Science Direct. Review methods: A comprehensive literature search was conducted without any restrictions on publication year, employing Boolean search techniques. The selection and review process of the studies followed PRISMA guidelines. The inclusion criteria comprised studies with participants aged 18 years and older who had nasal obstruction evaluated using acoustic voice analysis parameters, along with objective and/or subjective methods for assessing nasal obstruction. Results: Of the 174 abstracts identified, 118 were screened after the removal of duplicates. The full texts of 37 articles were reviewed. Only 10 studies met inclusion criteria. The majority of these studies found no significant correlations between voice parameters and nasal obstruction. Among the various acoustic parameters examined, shimmer was the most consistently affected, with statistically significant changes identified in three independent studies. A smaller number of studies reported notable findings for fundamental frequency (F0) and noise-related measures such as NHR/HNR. Conclusion: This systematic review critically evaluates existing studies on the use of voice analysis for assessing and monitoring nasal obstruction and hyponasality. The current evidence remains limited, as most investigations predominantly focus on glottic sound and dysphonia, with insufficient attention to the influence of the vocal tract, particularly the nasal cavities, on voice production. A notable gap exists in the integration of advanced analytical approaches, such as machine learning, in this field. Future research should focus on the use of advanced analytical approaches to specifically extrapolate the contribution of nasal resonance to voice thus defining the specific parameters in the voice spectrogram that can give precise information on nasal obstruction. Full article

Maintaining the required relative humidity values in the vehicle cabin is an important HVAC task, along with considerations related to the temperature, velocity, air pressure and noise. Deviation from the optimal values worsens the psycho-physiological state of the driver and affects the energy efficiency of the train. In this study, a model of liquid film formation on and removal from various cabin surfaces was constructed using the fundamental Navier–Stokes hydrodynamic equations. A special transport model based on the liquid vapor diffusion equation was used to simulate the air environment inside the cabin. The evaporation and condensation of surface films were simulated using the Euler film model, which directly considers liquid–gas and gas–liquid transitions. Numerical results were obtained using the RANS equations and a turbulence model by means of the finite volume method in Ansys CFD. Conjugate fields of temperature, velocity and moisture concentration were constructed for various time intervals, and the dependence values for the film thicknesses on various surfaces relative to time were determined. The verification was conducted in comparison with the experimental data, based on the protocol for measuring the microclimate indicators in workplaces, as applied to the train cabin: the average ranges encompassed temperature changes from 11% to 18%, and relative humidity ranges from 16% to 26%. Comparison with the results of other studies, without considering the phase transition and condensation, shows that, for the warm mode, the average air temperature in the cabin with condensation is 12.5% lower than without condensation, which is related to the process of liquid evaporation from the heated walls. The difference in temperature values for the model with and without condensation ranged from −12.5% to +4.9%. We demonstrate that, with an effective mode of removing condensate film from the window surface, including recirculation modes, the energy consumption of the climate control system improves significantly, but this requires a more accurate consideration of thermodynamic parameters and relative humidity. Thus, considering the moisture condensation model reveals that this variable can significantly affect other parameters of the microclimate in cabins: in particular, the temperature. This means that it should be considered in the numerical modeling, along with the basic heat transfer equations. Full article

Ensuring data integrity and adaptability in aircraft health monitoring (AHM) is vital for safety-critical aviation systems. Traditional digital twin (DT) and federated learning (FL) frameworks, while effective in enabling distributed, privacy-preserving fault detection, lack mechanisms to remove the influence of corrupted or adversarial data once these have been integrated into global models. This paper proposes a novel FL–DT–FU framework that combines digital twin-based subsystem modeling, federated learning for collaborative training, and federated unlearning (FU) to support the post hoc correction of compromised model contributions. The architecture enables real-time monitoring through local DTs, secure model aggregation via FL, and targeted rollback using gradient subtraction, re-aggregation, or constrained retraining. A comprehensive simulation environment is developed to assess the impact of sensor drift, label noise, and adversarial updates across a federated fleet of aircraft. The experimental results demonstrate that FU methods restore up to 95% of model accuracy degraded by data corruption, significantly reducing false negative rates in early fault detection. The proposed system further supports auditability through cryptographic logging, aligning with aviation regulatory standards. This study establishes federated unlearning as a critical enabler for resilient, correctable, and trustworthy AI in next-generation AHM systems. Full article

The implementation of an Indoor Positioning System (IPS) in a three-storey residential building employing WiFi signals that can also be used to track indoor movements is presented in this study. The movement of inhabitants is monitored through an Android smartphone by detecting the Received Signal Strength Indicator (RSSI) signals from WiFi Anchor Points (APs).Indoor movement is detected through a successive estimation of a target’s multiple positions. Using the K-Nearest Neighbors (KNN) and Particle Swarm Optimization (PSO) algorithms, these RSSI measurements are trained for estimating the position of an indoor target. Additionally, the Density-based Spatial Clustering of Applications with Noise (DBSCAN) has been integrated into the PSO method for removing RSSI-estimated position outliers of the mobile device to further improve indoor position detection and monitoring accuracy. We also employed Time Reversal Resonating Strength (TRRS) as a correlation technique as the third method of localization. Our extensive and rigorous experimentation covers the influence of various weather conditions in indoor detection. Our proposed localization methods have maximum accuracies of 92%, 80%, and 75% for TRRS, KNN, and PSO + DBSCAN, respectively. Each method also has an approximate one-meter deviation, which is a short distance from our targets. Full article

The efficient and reliable maintenance and repair of industrial machinery depend critically on circuit diagrams, which serve as essential references for troubleshooting and must be updated when machinery is modified. However, many circuit diagrams are not available in structured, machine-readable format; instead, they often exist as unstructured PDF files, rendered images, or even photographs. Existing digitization methods often address isolated tasks, such as symbol detection, but fail to provide a comprehensive solution. This paper presents a novel pipeline for extracting the underlying graph structures of circuit diagrams, integrating image preprocessing, pattern matching, and graph extraction. A U-net model is employed for noise removal, followed by gray-box pattern matching for device classification, line detection by morphological operations, and a final graph extraction step to reconstruct circuit connectivity. A detailed error analysis highlights the strengths and limitations of each pipeline component. On a skewed test diagram from a scan with slight rotation, the proposed pipeline achieved a device detection accuracy of 88.46% with no false positives and a line detection accuracy of 94.7%. Full article

Beef fat injection technology, used to enhance the perceived quality of lower-grade meat, often results in artificially marbled beef that mimics the visual traits of Wagyu, characterized by dense fat distribution. This practice, driven by the high cost of Wagyu and the affordability of fat-injected beef, has led to the proliferation of mislabeled “Wagyu-grade” products sold at premium prices, posing potential food safety risks such as allergen exposure or consumption of unverified additives, which can adversely affect consumer health. Addressing this, this study introduces a smart sensing system integrated with handheld mobile devices, enabling consumers to capture beef images during purchase for real-time health-focused assessment. The system analyzes surface texture and color, transmitting data to a server for classification to determine if the beef is artificially marbled, thus supporting informed dietary choices and reducing health risks. Images are processed by applying a region of interest (ROI) mask to remove background noise, followed by partitioning into grid blocks. Local binary pattern (LBP) texture features and RGB color features are extracted from these blocks to characterize surface properties of three beef types (Wagyu, regular, and fat-injected). A support vector machine (SVM) model classifies the blocks, with the final image classification determined via majority voting. Experimental results reveal that the system achieves a recall rate of 95.00% for fat-injected beef, a misjudgment rate of 1.67% for non-fat-injected beef, a correct classification rate (CR) of 93.89%, and an F1-score of 95.80%, demonstrating its potential as a human-centered healthcare tool for ensuring food safety and transparency. Full article

This paper presents a methodology for extracting and simulating arterial pulse waveform signals from Fabry–Perot interferometric measurements, emphasizing a practical approach for noninvasive cardiovascular assessment. A key novelty of this work is the presentation of a complete Python-based processing pipeline, which is made publicly available as open-source code on GitHub (git version 2.39.5). To the authors’ knowledge, no such repository for demodulating these specific interferometric signals to obtain a raw arterial pulse waveform previously existed. The proposed system utilizes accessible Python-based preprocessing steps, including outlier removal, Butterworth high-pass filtering, and min–max normalization, designed for robust signal quality even in settings with common physiological artifacts. Key features such as the rate of change, the Hilbert transform of the rate of change (envelope), and detected extrema guide the signal reconstruction, offering a computationally efficient pathway to reveal its periodic and phase-dependent dynamics. Visual analyses highlight amplitude variations and residual noise sources, primarily attributed to sensor bandwidth limitations and interpolation methods, considerations critical for real-world deployment. Despite these practical challenges, the reconstructed arterial pulse waveform signals provide valuable insights into arterial motion, with the methodology’s performance validated on measurements from three subjects against synchronized ECG recordings. This demonstrates the viability of Fabry–Perot sensors as a potentially cost-effective and readily implementable tool for noninvasive cardiovascular diagnostics. The results underscore the importance of precise yet practical signal processing techniques and pave the way for further improvements in interferometric sensing, bio-signal analysis, and their translation into clinical practice. Full article

Using interferometric synthetic aperture radar (InSAR) to observe slow ground deformation can be challenging due to many sources of error, with tropospheric phase delay and unwrapping errors being the most significant. While analytical methods, weather models, and data exist to mitigate tropospheric error, most of these techniques are unsuitable for all InSAR applications (e.g., complex tropospheric mixing in the tropics) or are deficient in spatial or temporal resolution. Likewise, there are methods for removing the unwrapping error, but they cannot resolve the true phase when there is a high prevalence (>40%) of unwrapping error in a set of interferograms. Applying tropospheric delay removal techniques is unnecessary for C-band Sentinel-1 InSAR time series studies, and the effect of unwrapping error can be minimized if the full dataset is utilized. We demonstrate that using interferograms with long temporal baselines (800 days to 1600 days) but very short perpendicular baselines (<5 m) (LTSPB) can lower the velocity detection threshold to 2 mm y⁻¹ to 3 mm y⁻¹ for long-term coherent permanent scatterers. The LTSPB interferograms can measure slow deformation rates because the expected differential phases are larger than those of small baselines and potentially exceed the typical noise amplitude while also reducing the sensitivity of the time series estimation to the noise sources. The method takes advantage of the Sentinel-1 mission length (2016 to present), which, for most regions, can yield up to 300 interferograms that meet the LTSPB baseline criteria. We demonstrate that low velocity detection can be achieved by comparing the expected LTSPB differential phase measurements to synthetic tests and tropospheric delay from the Global Navigation Satellite System. We then characterize the slow (~3 mm/y) ground deformation of the Socorro Magma Body, New Mexico, and the Tampa Bay Area using LTSPB InSAR analysis. The method we describe has implications for simplifying the InSAR time series processing chain and enhancing the velocity detection threshold. Full article

The proliferation of social media platforms has triggered an unprecedented increase in multimodal fake news, creating pressing challenges for content authenticity verification. Current fake news detection systems predominantly rely on isolated unimodal analysis (text or image), failing to exploit critical cross-modal correlations or leverage latent social context cues. To bridge this gap, we introduce the SCCN (Semantic-enhanced Cross-modal Co-attention Network), a novel framework that synergistically combines multimodal features with refined social graph signals. Our approach innovatively combines text, image, and social relation features through a hierarchical fusion framework. First, we extract modality-specific features and enhance semantics by identifying entities in both text and visual data. Second, an improved co-attention mechanism selectively integrates social relations while removing irrelevant connections to reduce noise and exploring latent informative links. Finally, the model is optimized via cross-entropy loss with entropy minimization. Experimental results for benchmark datasets (PHEME and Weibo) show that SCCN consistently outperforms existing approaches, achieving relative accuracy enhancements of 1.7% and 1.6% over the best-performing baseline methods in each dataset. Full article

The future of the automotive industry appears to hinge on the integration of dissimilar materials, such as aluminum alloys and carbon steel. However, this combination can lead to galvanic corrosion, compromising the structural integrity. In this study, laser-welded joints of 6013-T4 aluminum alloy and DP980 steel were evaluated for their morphology, microhardness, and corrosion resistance. Corrosion resistance was assessed using the electrochemical noise technique over time in 0.1 M Na₂SO₄ and 3.5% NaCl solutions. The wavelet function was applied to remove the DC trend, and energy diagrams were generated to identify the type of corrosive process occurring on the electrodes. Corrosion on the electrodes was also monitored using photomicrographic images. Analysis revealed an aluminum–steel mixture in the melting zone, along with the presence of AlFe, AlFe₃, and AlI₃Fe₄ intermetallic compounds. The highest Vickers microhardness was observed in the heat-affected zone, adjacent to the melt zone, where a martensitic microstructure was identified. The 6013-T4 aluminum alloy demonstrated the highest corrosion resistance in both media. Conversely, the electrochemical noise resistance was similar for the DP980 steel and the weld bead, indicating that the laser welding process does not significantly impact this property. The energy diagrams showed that localized pitting corrosion was the predominant form of corrosion. However, generalized and mixed corrosion were also observed, which corroborated the macroscopic analysis of the electrodes. Full article

Wearable laser Doppler devices are susceptible to complex noise interferences, such as Gaussian white noise, baseline drift, and motion artifacts, with motion artifacts significantly impacting clinical diagnostic accuracy. Addressing the limitations of existing denoising methods—including traditional adaptive filtering that relies on prior noise information, modal decomposition techniques that depend on empirical parameter optimization and are prone to modal aliasing, wavelet threshold functions that struggle to balance signal preservation with smoothness, and the high computational complexity of deep learning approaches—this paper proposes an ISSA-VMD-AWPTD denoising algorithm. This innovative approach integrates an improved sparrow search algorithm (ISSA), variational mode decomposition (VMD), and adaptive wavelet packet threshold denoising (AWPTD). The ISSA is enhanced through cubic chaotic mapping, butterfly optimization, and sine–cosine search strategies, targeting the minimization of the envelope entropy of modal components for adaptive optimization of VMD’s decomposition levels and penalty factors. A correlation coefficient-based selection mechanism is employed to separate target and mixed modes effectively, allowing for the efficient removal of noise components. Additionally, an exponential adaptive threshold function is introduced, combining wavelet packet node energy proportion analysis to achieve efficient signal reconstruction. By leveraging the rapid convergence property of ISSA (completing parameter optimization within five iterations), the computational load of traditional VMD is reduced while maintaining the denoising accuracy. Experimental results demonstrate that for a 200 Hz test signal, the proposed algorithm achieves a signal-to-noise ratio (SNR) of 24.47 dB, an improvement of 18.8% over the VMD method (20.63 dB), and a root-mean-square-error (RMSE) of 0.0023, a reduction of 69.3% compared to the VMD method (0.0075). The processing results for measured human blood flow signals achieve an SNR of 24.11 dB, a RMSE of 0.0023, and a correlation coefficient (R) of 0.92, all outperforming other algorithms, such as VMD and WPTD. This study effectively addresses issues related to parameter sensitivity and incomplete noise separation in traditional methods, providing a high-precision and low-complexity real-time signal processing solution for wearable devices. However, the parameter optimization still needs improvement when dealing with large datasets. Full article

Accurately predicting recycling prices at battery recycling sites helps reduce transportation and dismantling costs, ensures economies of scale in the recycling, and supports the sustainable development of the new energy vehicle industry. However, this prediction typically relies on easily accessible surface data, such as battery characteristics and market prices. These data have complex correlations with recycling price, general price prediction methods have low prediction accuracy. To this end, an improved prediction method is proposed to enhance the accuracy of predicting recycling prices through surface data. Firstly, factors influencing recycling prices are selected based on self-factor and market fluctuations, a bidirectional denoising autoencoder and support vector regression model (BDAE-SVR) is established. BDAE is used to adjust the weights of influencing factors to remove noise, extract features related to recycling price. The extracted features are introduced into the SVR model to establish a correspondence between the features and recycling price. Secondly, to have better applications for different batteries, the Grey Wolf algorithm (GWO) is used to adjust the SVR parameters to improve the generalization ability of the prediction model. Finally, taking retired power batteries as an example, the effectiveness of the method is verified. Compared with methods such as random forest (RF), the RMSE predicted by BDAE is decreased from 1.058 to 0.371, indicating better prediction accuracy. Full article