Search Results (7,988)

Rolling element bearings are critical components in rotating machinery, yet assessing and predicting their reliability under heavy industrial noise remains challenging. Existing methods suffer from three major limitations: (1) single-channel signal processing and single-scale indicators lack robustness against non-stationary noise; (2) classical multi-channel decomposition methods, such as multivariate variational mode decomposition (MVMD), rely on empirical parameter tuning, which frequently leads to over- or under-decomposition; and (3) monolithic deep architectures and homogeneous ensemble models suffer from prediction drift and generalization bottlenecks during long-term temporal extrapolation. To address these issues, this paper introduces an automated framework that combines adaptive multi-channel signal purification with a heterogeneous stacking ensemble (HeteroStack-LR). Unlike conventional MVMD pipelines that fix $[K,\alpha ]$ empirically, the Sequoia Optimization Algorithm (SOA) autonomously determines the globally optimal configuration, achieving a mean SNR of $-2.08\mathrm{dB}$—a $1.88$ to $2.50\mathrm{dB}$ improvement over standard VMD/MVMD baselines—along with up to a $37.1\%$ reduction in computation time. Rather than relying on conventional single-metric intrinsic mode function (IMF) selection, we construct a multi-domain hybrid index integrating the Fault Correlation Factor, energy ratio, and refined composite multiscale dispersion entropy (RCMDE) to robustly identify noise-resistant components, thereby enhancing denoising quality by $22.4\%$ to $32.2\%$ over single-scale criteria. Furthermore, contrasting with linear PCA-based reduction, Diffusive Topology Neighbor Embedding (D-TNE) effectively preserves the nonlinear manifold structure of degradation trajectories in a low-dimensional space. Finally, a heterogeneous stacked ensemble featuring an out-of-fold (OOF) leakage-prevention strategy and a logistic regression meta-learner is designed to suppress prediction drift while avoiding the over-parameterization typical of deep architectures. Experimental results across four bearing datasets demonstrate that HeteroStack-LR achieves a minimal MAE of $0.063$ with a variance of ≤±0.002, outperforming state-of-the-art deep architectures (such as TCN, CNN-LSTM, BiLSTM-Attention, and Transformer) as well as classical baselines (Bi-LSTM, CNN, and LSSVM). Ablation studies confirm that removing SOA and MVMD degrades MAE by $12.7\%$ and $19.0\%$, respectively, validating that the framework’s strength stems not from any isolated module, but from the end-to-end synergistic integration of signal purification and reliability prediction. Full article

Objectives: The objective of this study was to evaluate the impact of virtual monoenergetic image (VMI) reconstructions derived from photon-counting computed tomography (PCCT) on the assessment of carotid arteries, with a focus on optimizing keV selection based on plaque composition. Methods: This retrospective study included 111 patients (mean age 80 ± 7.5 years; 64 men; 47 women) with carotid sclerosis who underwent PCCT between April 2022 and February 2023. One lesion was analyzed per patient, each containing both calcified and non-calcified components. Quantitative measurements were performed in calcified plaque across energy levels from 40 to 120 keV and comprised attenuation, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), corrected image noise (CIN) and artifact index (AIX). Two radiologists independently rated image quality, artifacts, and diagnostic assessability of both components using five-point scales. Results: Attenuation and CNR were highest at 40 to 50 keV, where CIN and AIX were also greatest. SNR followed a U-shaped course, lowest at 90 keV and highest at 110 to 120 keV, where intraluminal attenuation was too low for reliable luminal delineation. Reader ratings were highest at 40 to 50 keV for non-calcified components and at 70 to 80 keV for calcified plaque, with good interobserver agreement throughout (κ 0.74 to 0.92). After correction for multiple testing, adjacent energy levels were frequently indistinguishable. Conclusions: Optimal PCCT VMI energy levels depend on plaque composition. The findings support implementing standardized protocols with automatic dual-range reconstructions (40–50 keV and 70–80 keV) to enable efficient, individualized carotid plaque assessment in clinical practice. Full article

Industrial IoT sensor nodes face a triple burden—sampling, compression, and security—under severe resource constraints; yet, the question of which signals can actually benefit from compressed sensing (CS) remains largely implicit in the literature. SCCS answers this question by unifying compression, chaotic obfuscation, and authentication within a single CS measurement and deriving an empirical deployability rule consisting of the PCA energy concentration ratio $\rho$. When $\rho$ exceeds 80%, signals reconstruct at high fidelity; when $\rho$ falls below 50%, they are intrinsically incompressible; and in the intermediate 50–80% band, reconstruction is uncertain and may fail outright rather than degrading gracefully (as shown on CWRU). This empirical deployability rule is supported by evaluation on three real datasets: high-fidelity reconstruction is confirmed on CBM ($\rho =99.9\%$), while CWRU ($\rho =65.9\%$) and CCPP ($\rho =15.4\%$) establish the applicability boundaries and validate the $\rho$-based screening criterion. The enabling system integrates a block-circulant chaotic measurement matrix (BCCM, from a two-dimensional sine-logistic iteration mapping (2D-SLIM) map) that compresses and obfuscates in one operation (online measurement seed 0.84 KB, down from a 512 KB dense matrix; the full reference implementation requires 185 KB Flash, including a 160 KB decoder dictionary); an offline principal component analysis (PCA) dictionary that lifts reconstruction signal-to-noise ratio (SNR) from 5.36 to 33.52 dB at CR = 4 (+28.16 dB over the fixed-basis configuration; Wilcoxon $p<0.001$, 30 independent trials); and a dual-layer authentication scheme combining always-on hash-based message authentication code (HMAC) with adaptive reconstruction-based implicit authentication (RBIA), the latter providing zero-overhead tamper pre-screening that reuses the decoder’s reconstruction residual and automatically falls back to HMAC-only under channel noise. Security boundaries are explicitly disclosed: the chaotic measurement resists known-plaintext attacks but is vulnerable to chosen-plaintext recovery (N plaintexts recover the linear matrix), and 1.13 bits of amplitude side-channel leakage exist. The SCCS framework demonstrates that the three functions need not be separate serial stages, provided the target signals satisfy the $\rho$ screening rule. Full article

Accurate quantitative SPECT depends strongly on the reconstruction protocol, particularly for penalised-likelihood reconstruction algorithms, such as Q.Clear (BSREM). This study aimed to optimise Q.Clear reconstruction parameters for ¹²³I and ¹⁷⁷Lu imaging. SPECT images of a NEMA IEC PET Body phantom filled with ¹²³I and ¹⁷⁷Lu were acquired on a ring-shaped CZT camera and on a conventional Anger camera. The images were reconstructed using Q.Clear and OSEM and evaluated using quantitative image-quality metrics. Q.Clear was optimised for each radionuclide. The ¹²³I dataset was reconstructed using different combinations of the regularisation parameters $\beta$ and $\gamma$. The ¹⁷⁷Lu dataset was reconstructed using literature-reported parameters, and the best-performing reconstruction was fine-tuned. The best-performing Q.Clear reconstructions were compared with corresponding OSEM reconstructions. Overall, increasing $\beta$ reduced the recovery coefficients, background variability, and contrast recovery coefficients while increasing the signal-to-noise ratio. $\gamma$ had no appreciable influence on the metrics within the investigated range. The candidate optimised StarGuide/Q.Clear reconstructions outperformed the standard Symbia/OSEM reconstructions in terms of activity recovery, whereas Symbia/OSEM produced lower image noise for most of the analysed VOIs. This study identified settings associated with improved quantitative metrics within the evaluated parameter space, with potential implications for quantitative imaging requiring clinical validation. Nevertheless, multiple reconstruction parameter combinations remain unexplored. Full article

Wearable near-infrared spectroscopy (NIRS) enables non-invasive hemodynamic monitoring, yet conventional CW-NIRS systems suffer from limited temporal resolution, scalp-only measurement, and mandatory manual tuning to compensate for optical heterogeneity across subjects and sites. This work develops NIRSLINK, a modular cascaded wearable NIRS system for high-speed multi-site hemodynamic acquisition. Its flexible probes adopt spring-floating optics with a standardized 30 mm optode separation, integrating dual 735/850 nm LEDs, silicon photodiodes, and a two-stage closed-loop tuning algorithm. A single probe achieves a peak sampling rate of 3 kHz, and up to eight cascaded probes form 52 valid channels, with a signal-to-noise ratio (SNR) of 78.61 ± 7.03 dB and an optical dynamic range (DR) of 101.32 ± 12.41 dB. Phantom experiments verify its millisecond temporal resolution and high sensitivity to blood flow and hemoglobin variations. In vivo trials, including the Valsalva maneuver, forearm occlusion, and two-back cognitive tasks, demonstrate simultaneous recording of hemoglobin concentration shifts, pulse waveforms, and beat-to-beat pulse transit times (PTTs). NIRSLINK supports hemodynamic measurements across multiple anatomical locations, including the forehead, forearm, upper arm, and thigh, covering both cranial and peripheral body regions. The embedded auto-tuning module stabilizes signals from the forehead, forearm, and other regions within the optimal ADC range without manual adjustment, preventing signal saturation and SNR degradation. This scalable adaptive platform overcomes critical drawbacks of traditional wearable NIRS, applicable to cognitive neuroscience, non-invasive cardiovascular assessment, and ambulatory physiological monitoring, and provides design references for multi-site optical sensors. Full article

Reliable detection of event-related potentials (ERPs) at the single-trial level remains a challenge due to low signal-to-noise ratio and high variability in electroencephalography (EEG) recordings. This work investigates the use of the Deep-Match framework (Deep-MF) for ERP detection. We examine whether incorporating prior knowledge of an ERP template into deep learning models improves detection performance. As a proof-of-concept study, the framework was evaluated on a single dataset with multi-channel EEG recordings during laser stimulation. The model was trained in two stages. First, an encoder–decoder architecture was trained to reconstruct input EEG signals in order to learn compact signal representations. In the second stage, the decoder was replaced with a detection module and the network was fine-tuned for ERP identification. Two model variants were evaluated: a standard model with randomly initialized filters and a Deep-MF model in which input kernels were initialized using ERP templates. Models performance was assessed on a single-trial ERP detection task during leave-one-out validation, and then compared with matched filter detector. The neural network models outperformed the matched filter detector and proposed that the Deep-MF model slightly outperformed the detector with standard kernel initialization for the majority of held-out subjects. Although both approaches exhibited substantial inter-subject variability, Deep-MF achieved a higher average F1-score (0.37) compared to the standard network (0.34), indicating improved robustness to cross-subject differences. Performance varied considerably across participants. The best performance obtained by Deep-MF reached an F1-score of 0.71, exceeding the maximum score achieved by the standard model (0.59). These results showed that ERP-informed kernel initialization provides improvements in single-trial ERP detection under subject-independent evaluation. These findings demonstrate that integrating domain knowledge with deep learning architectures can improve single-trial ERP detection. The proposed approach provides a step towards practical wearable EEG and passive brain–computer interface applications, as well as towards real-time monitoring of cognitive processes. Full article

This study focuses on accelerating adaptive beamforming in the Multicarrier Wireless Internet Local Loop (McWiLL) professional radio communication system using “Korona” ring smart antennas. The work investigates algorithms for calculating complex weight coefficients in an eight-element uniform circular antenna array. The main objective is to reduce beam pattern adaptation time while maintaining interference suppression depth and robustness under multipath propagation. To achieve this, an ensemble machine learning approach based on the Random Forest algorithm is employed to approximate the optimal Minimum Variance Distortionless Response (MVDR) solution using elements of the sample covariance matrix of received signals. The training dataset is generated through McWiLL channel simulations considering mutual coupling between array elements, signal-to-noise ratio (SNR) variation, and different angles of arrival of the desired and interfering signals. The proposed method is evaluated against the classical MVDR algorithm in terms of radiation pattern null depth, robustness to phase distortions, and inference time on a Field-Programmable Gate Array (FPGA) hardware platform. Results demonstrate that the Random Forest-based approach achieves more than a fourfold reduction in computation time while forming radiation-pattern nulls of about 30–35 dB toward the interferers (versus 44–46 dB for the classical MVDR); the synthesized core uses no hardware multipliers (DSP48), and its functional equivalence to the software model is confirmed by bit-exact RTL co-simulation. The findings show promise for deployment in McWiLL base stations and other professional radio systems requiring fast, adaptive beamforming under dynamic channel conditions. Full article

Integrating uncrewed aerial vehicles (UAVs) and high-altitude platform stations (HAPS) into terrestrial cellular wireless networks is central to emerging sixth-generation (6G) communication systems. However, the energy budgets of both the ground network and the power-constrained aerial platforms are a serious concern. Cell switching, which selectively places lightly loaded small base stations (SBSs) into sleep mode, is an important energy-saving mechanism, but existing feasibility criteria are typically based either on maintaining a target data rate or merely preserving coverage for the users originally served by the switched-off SBSs (i.e., displaced users). These two approaches represent opposite ends of the quality-energy tradeoff: rate-based policies require a high signal-to-interference-plus-noise ratio (SINR), limiting energy savings, whereas coverage-based policies permit more aggressive sleeping at the expense of quality of service (QoS). This work proposes a novel semantic-aware cell switching policy whose feasibility criterion is a semantic-service requirement rather than a bit-centric target: an SBS is put into a sleep mode if and only if its displaced users still satisfy an assumed semantic-service SINR criterion. The policy is tested in a multi-tier heterogeneous network comprising an always-on macro base station (MBS), switchable SBSs, a tier of UAV base stations (UAV-BSs), and a HAPS-mounted International Mobile Telecommunications (IMT) base station (HIBS), and it is solved by a greedy algorithm. For the deployment and parameter set considered, and with the HIBS present, the semantic criterion deactivates the entire SBS layer while every user continues to satisfy the assumed semantic-service SINR criterion, whereas the rate criterion deactivates none. Full article

The rapid development of Industry 4.0 brings significant benefits to intralogistics through increased flexibility, interoperability, and real-time responsiveness. However, the growing complexity of industrial environments creates challenges for logistics process adaptation and the integration of emerging technologies such as voice-driven control of heterogeneous robotic and IoT systems. Open-source solutions offer a promising approach by enabling vendor independence and cost-effective deployment. This paper investigates the use of the OPIL cyber–physical middleware, developed within the Horizon 2020 L4MS initiative, combined with convolutional neural networks for speech denoising and keyword recognition, to enable robust voice-controlled intralogistics in demanding noisy industrial environments. A voice-controlled logistics system was designed, deployed, and empirically evaluated on the OPIL platform. The results confirm the technical feasibility of OPIL as an open and modular alternative to proprietary Industry 4.0 platforms. They further demonstrate the feasibility of integrating AI-based keyword spotting and voice interaction into environments using an open middleware architecture. The study also identifies key limitations, including reduced performance for non-native speakers and mild speech over-suppression at high signal-to-noise ratios. To support reproducibility and external validation, the complete network configurations, training procedure, and deployment data are provided. Full article

Objective: Recent advances in dry electrode EEG have enabled rapid setup and recording in unconventional scenarios. However, past developments were primarily driven by brain–computer interfaces (BCI), leaving their comparability to wet electrodes in clinical and daily life applications an open question. Here, we developed a new dry EEG system and systematically benchmarked its performance against a commercial wet EEG system across various tasks. Methods: Participants (n = 19) underwent simultaneous recording using both devices. We first collected resting-state EEG under both eyes-closed and eyes-open conditions, followed by a steady-state visual evoked potential (SSVEP) task at different flicker frequencies and a motor imagery (MI) task. System performance was evaluated using power spectral density (PSD), signal to noise ratio (SNR), event-related spectral perturbation (ERSP), and single-trial classification accuracy. Results: The two systems performed similarly across different tasks. During the resting state, no statistically significant differences were observed between the two systems in the PSD of the five frequency bands (p > 0.05 in all cases). Similarly, SNR in the SSVEP task showed no significant differences at 8 Hz, 10 Hz, and 12 Hz after correction. For cognitive tasks, classification accuracies were comparable (SSVEP: dry 80.08% ± 7.1% vs. wet 81.10% ± 6.5%; MI: dry 72.46% ± 3.89% vs. wet 70.7% ± 2.37%). Conclusions: The developed dry EEG system can effectively record electrophysiological measurements commonly employed in research and clinical settings, with quality comparable to that of traditional wet EEG systems. Full article

Visible–near-infrared (VNIR) hyperspectral imaging provides a non-contact approach for cultural heritage examination. This study presents the design and calibration of a compact finite-conjugate VNIR pushbroom hyperspectral camera for close-range mural imaging. Operating over 400–1000 nm at a nominal working distance of 404 mm, the system provides a mean spectral sampling interval of 4.85 nm and an object-space sampling interval of approximately 82.4 μm/pixel. An integrated calibration workflow was established for wavelength assignment, spectral response characterization, geometric correction, radiometric calibration, and scan synchronization. The experimental results yielded a modulation transfer function (MTF) of 0.34 at the effective detector Nyquist frequency, a mean spectral response function full width at half maximum (FWHM) of 6.3 nm, a maximum absolute wavelength residual below 0.90 nm, residual smile and keystone errors below 0.3 pixels, a residual radiometric nonuniformity of 0.71%, and a mean signal-to-noise ratio (SNR) of 339. Measurements of Potala Palace mural samples demonstrate the acquisition of spatially detailed, radiometrically corrected hyperspectral data under close-range conditions. Full article

The electromagnetic spectrum grows increasingly crowded with the rapid expansion of mobile, satellite and Internet of Things communications, making intelligent spectrum sensing and efficient management an urgent priority. Automatic modulation classification (AMC) serves as the core of cognitive radio and intelligent communication. Existing deep models often suffer from a large number of parameters and low storage efficiency. To overcome these limitations, we propose MDSCNet, a multi-scale depth-wise separable complex network. Built upon complex depth-wise separable convolution, the network makes full use of the phase information in in-phase and quadrature signals while naturally preserving the symmetric relationship between the in-phase and quadrature components (IQ). The asymmetric multi-scale structure combined with the embedded lightweight attention module jointly forms the overall feature extraction process. The overall parameter count is kept extremely low, at only 47.739 k. Experiments on the RML2016.10a and RML2016.10b datasets show that MDSCNet delivers recognition performance under low signal-to-noise ratios (SNR), reaching 63.42% and 66.71% respectively. More importantly, it outperforms mainstream methods in both parameter count and storage efficiency. Full article

Computational imaging shifts part of the aberration correction from optical hardware to algorithms, offering a viable path toward compact, simplified systems. However, overexposed regions—often caused by phenomena such as water-body reflections—can readily induce severe ringing artifacts in reconstructed images. To address this problem, we propose a Ringing-perceptive Cooperative Reconstruction Network (RPCR-Net). This network integrates a learned Wiener filter and a field-of-view shared kernel prediction network (FOV-KPN) for feature extraction and innovatively incorporates a combined regularization mechanism that leverages a Local Maximum Gradient Prior and a multi-scale ringing measurement model within its loss function to suppress artifacts while preserving details. Validated on a constructed overexposed image dataset, RPCR-Net improves the Peak Signal-to-Noise Ratio (PSNR) from 29.08 dB to 37.06 dB and the Structural Similarity Index Measure (SSIM) from 0.8795 to 0.9549. Experiments on real-world scenes further confirm its capability to suppress ringing artifacts while maintaining visual quality. The proposed method can generate high-quality images such as image reconstruction and robustness improvement in optical systems. Full article

Infrared small target detection (IRSTD) is a challenging task in intelligent infrared sensing and electronic imaging systems, because dim targets often occupy only a few pixels and are easily disturbed by clutter, noise, and low-contrast background structures. A practical detector should preserve pixel-level target cues while suppressing target-like false responses. This paper proposes a Pyramid Target Perception Network (PTPN) for single-frame pixel-level IRSTD. The network integrates three complementary components: an Efficient Context Modeling (ECM) encoder employing 7 × 7 depthwise separable convolution for lightweight contextual feature extraction, a multi-scale target cross-attention (MTCA) module for hierarchical feature interaction, and a small-target feature pyramid network (STFPN) for target-preserving multi-scale aggregation. In addition, a physics-constrained loss (PCL) is introduced during training to regularize predictions according to infrared imaging characteristics, including point spread consistency, target-region relative intensity consistency, and signal-to-noise-ratio-aware separability. Experiments on IRSTD-1k, NUAA-SIRST, and NUDT-SIRST demonstrate that PTPN achieves $IoU$ scores of 71.87%, 79.56%, and 86.47%, respectively, with 4.55M parameters, 4.96G FLOPs at an input resolution of 256 × 256, and an inference speed of 45.0 FPS. Although PTPN achieves competitive overall performance, it does not attain the highest $IoU$ on NUDT-SIRST, indicating that pixel-level target-region estimation under complex scenes remains an area for further improvement. Overall, PTPN provides an effective balance between target localization, false-alarm suppression, and computational efficiency, supporting its potential application in AI-driven infrared image processing and intelligent electronic sensing systems. Full article

Accurate detection of sub-millimeter defects in reactor core-plate cotter-pin holes is essential for nuclear safety. However, underwater inspection images often suffer from low signal-to-noise ratios, weak boundary responses, and pseudo-edge interference, resulting in unstable localization of defects. Existing deformable and attention-based detectors remain vulnerable to sampling drift and semantic–boundary inconsistency under such conditions. To address these challenges, an Edge-Geometry-Guided Deformable Detection Network (EGD-Net) is proposed for underwater defect detection. EGD-Net introduces an edge-geometry-constrained deformable sampling mechanism that embeds edge-confidence priors into deformable convolution to improve boundary-aware feature sampling. A cross-level semantic–geometric alignment strategy is designed to enhance the interaction between defect semantics and geometric boundary cues, while a top-down feedback recalibration mechanism improves multi-scale response consistency for weak defects. Experiments on the Core-Plate Pin-Hole Defect (CPHD) dataset demonstrate that EGD-Net achieves the highest AP@[0.5:0.95] on both datasets while maintaining competitive or superior Precision, Recall, and F1-score while reducing engineering center error under a fixed operating point. Performance across the two complementary domains suggests its robustness to variations between coupon images and practical underwater inspection scenes. These results indicate that EGD-Net provides a reliable solution for boundary-sensitive localization of underwater sub-millimeter defects in nuclear inspection. Full article