Search Results (9,092)

This study presents a segmentation framework for images of 220 kV cable insulation that addresses sample scarcity and blurred boundaries. The framework integrates data augmentation using the Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) and the TransUNet architecture. Considering the difficulty and high cost of obtaining real cable images, WGAN-GP generates high-quality synthetic data to expand the dataset and improve the model’s generalization. The TransUNet network, designed to handle the structural complexity and indistinct edge features of insulation layers, combines the local feature extraction capability of convolutional neural networks (CNNs) with the global context modeling strength of Transformers. This combination enables accurate delineation of the insulation regions. The experimental results show that the proposed method achieves mDice, mIoU, MP, and mRecall scores of 0.9835, 0.9677, 0.9840, and 0.9831, respectively, with improvements of approximately 2.03%, 3.05%, 2.08%, and 1.98% over a UNet baseline. Overall, the proposed approach outperforms UNet, Swin-UNet, and Attention-UNet, confirming its effectiveness in delineating 220 kV cable insulation layers under complex structural and data-limited conditions. Full article

The increasing elderly population presents major challenges to traditional healthcare due to the need for continuous care, a shortage of skilled professionals, and increasing medical costs. To address this, smart elderly care homes where multiple residents live with the support of caregivers and IoT-based assistive technologies have emerged as a promising solution. For their effective operation, a reliable high speed network like 5G is essential, along with intelligent resource allocation to ensure efficient service delivery. This study proposes a deep reinforcement learning (DRL)-based resource management framework for smart elderly homes, formulated as a Markov decision process. The framework dynamically allocates computing and network resources in response to real-time application demands and system constraints. We implement and compare two DRL algorithms, emphasizing their strengths in optimizing edge utilization and throughput. System performance is evaluated across balanced, high-demand, and resource-constrained scenarios. The results demonstrate that the proposed DRL approach effectively learns adaptive resource management policies, making it a promising solution for next-generation intelligent elderly care environments. Full article

This work presents a generative design algorithm for the semi-automatic reconstruction of sweepable surfaces from point clouds obtained through three-dimensional scanning. The proposed algorithm enables, starting from a 3D acquisition dataset, the correct automatic orientation of the mesh, the selection of a suitable cutting edge, and the specification of the number of transversal sections for an effective 3D model reconstruction. Additionally, it suggests a maximum number of points to be used for reconstructing the sectional curves. The mesh reconstruction is performed through a lofting operation, resulting in a non-uniform rational B-spline (NURBS) surface. The algorithm has been applied to a case study involving the front wing surface of a foil from the Olympic class iQFoil, which has recently garnered significant attention from researchers in the field of performance analysis. The obtained reconstructed surface exhibits very low deviation values when compared to the original mesh. This demonstrates the reliability of the results obtained with the proposed approach, which provides sufficient accuracy and is obtained in a considerably shorter time compared to the traditional manual reconstruction approach, enabling the reconstruction of a 3D model in just a few semi-automatic steps, ready for subsequent numerical analyses if needed. Full article

This study, inspired by the impact resistance of the pistol shrimp’s predatory claw, investigates the design and optimization of bionic energy absorption structures. Four types of bionic hierarchical porous metamaterial lattice structures with a negative Poisson’s ratio were developed based on the microstructure of the pistol shrimp’s fixed claw. These structures were validated through finite element models and quasi-static compression tests. Results showed that each structure exhibited distinct advantages and shortcomings in specific evaluation indices. To address these limitations, four new bionic structures were designed by coupling the characteristics of the original structures. The coupled structures demonstrated a superior balance across various performance indicators, with the EOS (Eight pillars Orthogonal with Side connectors on square frame) structure showing the most promising results. To further enhance the EOS structure, a parametric study was conducted on the distance d from the edge line to the curve vertex and the length-to-width ratio y of the negative Poisson’s ratio structure beam. A fifth-order polynomial surrogate model was constructed to predict the Specific Energy Absorption (SEA), Crush Force Efficiency (CFE), and Undulation of Load-Carrying fluctuation (ULC) of the EOS structure. A multi-objective genetic algorithm was employed to optimize these three key performance indicators, achieving improvements of 1.98% in SEA, 2.42% in CFE, and 2.05% in ULC. This study provides a theoretical basis for the development of high-performance biomimetic energy absorption structures and demonstrates the effectiveness of coupling design with optimization algorithms to enhance structural performance. Full article

In this study, a procedure based on Phase-locked Proper Orthogonal Decomposition (PPOD) was applied to Large Eddy Simulations (LESs) of two low-pressure turbine blades operating with unsteady inflow. This decomposition allows the inspection of the effect of blade loading on loss generation mechanisms, focusing especially on their variation throughout the incoming wake period. After sorting snapshots based on their phase within the wake cycle using temporal POD coefficients associated with wake migration, POD was reapplied to each sub-ensemble of snapshots at a given phase, providing an optimal representation of the dynamics at fixed wake locations. This highlighted the effects of the migration, bowing, tilting, and reorientation of the incoming wake filaments, as well as the breakup of streaky structures in the blade boundary layer and the formation of Von Karman vortices at the blade trailing edge. PPOD offered us the opportunity to observe how all these processes are modulated and change throughout the wake period. The comparison between the two analyzed blades showed that overall loss generation follows similar temporal patterns during the wake-passing cycle, increasing with the propagation of the upstream wake and reaching its maximum value when the wake is in the peak suction position. According to the specific blade loading distribution, the production of TKE was observed in different regions of the computational domain. The described procedure may contribute to the development of advanced design processes based on physically informed strategies. Full article

Precise and robust localization for micro Unmanned Aerial Vehicles (UAVs) in GNSS-denied environments is hindered by the lack of diverse datasets and the limited real-world performance of existing visual matching methods. To address these gaps, we introduce two contributions: (1) the SatLoc dataset, a new benchmark featuring synchronized, multi-source data from varied real-world scenarios tailored for UAV-to-satellite image matching, and (2) SatLoc-Fusion, a hierarchical localization framework. Our proposed pipeline integrates three complementary layers: absolute geo-localization via satellite imagery using DinoV2, high-frequency relative motion tracking from visual odometry with XFeat, and velocity estimation using optical flow. An adaptive fusion strategy dynamically weights the output of each layer based on real-time confidence metrics, ensuring an accurate and self-consistent state estimate. Deployed on a 6 TFLOPS edge computer, our system achieves real-time operation at over 2 Hz, with an absolute localization error below 15 m and effective trajectory coverage exceeding $90\%$, demonstrating state-of-the-art performance. The SatLoc dataset and fusion pipeline provide a robust and comprehensive baseline for advancing UAV navigation in challenging environments. Full article

Multimodal medical data provides a wide and real basis for disease diagnosis. Computer-aided diagnosis (CAD) powered by artificial intelligence (AI) is becoming increasingly prominent in disease diagnosis. CAD for multimodal medical data requires addressing the issues of data fusion and prediction. Traditionally, the prediction performance of CAD models has not been good enough due to the complicated dimensionality reduction. Therefore, this paper proposes a fusion and prediction model—EPGC—for multimodal medical data based on graph neural networks. Firstly, we select features from unstructured multimodal medical data and quantify them. Then, we transform the multimodal medical data into a graph data structure by establishing each patient as a node, and establishing edges based on the similarity of features between the patients. Normalization of data is also essential in this process. Finally, we build a node prediction model based on graph neural networks and predict the node classification, which predicts the patients’ diseases. The model is validated on two publicly available datasets of heart diseases. Compared to the existing models that typically involve dimensionality reduction, classification, or the establishment of complex deep learning networks, the proposed model achieves outstanding results with the experimental dataset. This demonstrates that the fusion and diagnosis of multimodal data can be effectively achieved without dimension reduction or intricate deep learning networks. We take pride in exploring unstructured multimodal medical data using deep learning and hope to make breakthroughs in various fields. Full article

In order to enhance the performance of 112 Gb/s polarization-multiplexed quadrature phase-shift keying (PM-QPSK) coherent optical receivers, a novel digital signal processing (DSP) framework is presented in this study. The suggested method combines cutting-edge signal processing techniques to address important constraints in long-distance, high data rate coherent systems. The framework uses overlap frequency domain equalization (OFDE) for chromatic dispersion (CD) compensation, which offers a cheaper computational cost and higher dispersion control precision than traditional time-domain equalization. An adaptive carrier phase recovery (CPR) technique based on mean-squared differential phase (MSDP) estimation is incorporated to manage phase noise induced by cross-phase modulation (XPM), providing dependable correction under a variety of operating situations. When combined, these techniques significantly increase Q factor performance, and optimum systems can handle transmission distances of up to 2400 km. The suggested DSP approach improves phase stability and dispersion tolerance even in the presence of nonlinear impairments, making it a viable and effective choice for contemporary coherent optical networks. The framework’s competitiveness was evaluated by comparing it against the most recent, cutting-edge DSP methods that were released after 2021. These included CPR systems that were based on kernels, transformers, and machine learning. The findings show that although AI-driven approaches had the highest absolute Q factors, they also required a large amount of computing power. On the other hand, the suggested OFDE in conjunction with adaptive CPR achieved Q factors of up to 11.7 dB over extended distances with a significantly reduced DSP effort, striking a good balance between performance and complexity. Its appropriateness for scalable, long-haul 112 Gb/s PM-QPSK systems is confirmed by a complexity versus performance trade-off analysis, providing a workable and efficient substitute for more resource-intensive alternatives. Full article

Unmanned Aerial Vehicles (UAVs) integrated with lightweight deep learning models represent an effective solution for image-based rapid post-earthquake damage assessment. UAVs, equipped with cameras, capture high-resolution aerial imagery of disaster-stricken areas, providing essential data for evaluating structural damage. When paired with light eight Convolutional Neural Network (CNN) models, these UAVs can process the captured images onboard, enabling real-time, accurate damage level predictions that might with potential interest to orient efficiently the efforts of the Search and Rescue (SAR) teams. This study investigates the use of the MobileNetV3-Small lightweight CNN model for real-time post-earthquake damage level prediction using UAV-captured imagery. The model is trained to classify three levels of post-earthquake damage, ranging from no damage to severe damage. Experimental results show that the adapted MobileNetV3-Small model achieves the lowest FLOPs, with a significant reduction of 58.8% compared to the ShuffleNetv2 model. Fine-tuning the last five layers resulted in a slight increase of approximately 0.2% in FLOPs, but significantly improved accuracy and robustness, yielding a 4.5% performance boost over the baseline. The model achieved a weighted average F-score of 0.93 on a merged dataset composed of three post-earthquake damage level datasets. It was successfully deployed and tested on a Raspberry Pi 5, demonstrating its feasibility for edge-device applications. This deployment highlighted the model’s efficiency and real-time performance in resource-constrained environments. Full article

While hierarchical subarray-based architectures are widely adopted for computational efficiency, the side effects of subarray management on the array factor and overall radiation behaviour remains insufficiently addressed. Array grid and perimeter misalignment with the subarrays can cause side effects such as increased sidelobe levels and distorted beam shapes. By implementing well-oriented, periodic subarray distributions, we show how one can achieve improved radiation performance, enhanced beam symmetry, and reduced sidelobe interference. We establish clear cause–effect relationships between subarray orientation, edge structure, and the resulting radiation patterns. Simulation results demonstrate that precise geometric alignment of subarrays leads to better energy concentration and consistent array factors. Our design approach is highly suitable for scalable, fully digital beamforming systems. Full article

Wireless sensor networks (WSNs) are fundamental to applications in the Internet of Things, smart cities, and environmental monitoring, where coverage optimization is critical for maximizing monitoring efficacy under constrained resources. Conventional approaches often suffer from low global coverage efficiency, high computational overhead, and a tendency to converge to local optima. To address these challenges, this study proposes the fusion multi-strategy gray wolf optimizer (FMGWO), an advanced variant of the Gray Wolf Optimizer (GWO). FMGWO integrates various strategies: electrostatic field initialization for uniform population distribution, dynamic parameter adjustment with nonlinear convergence and differential evolution scaling, an elder council mechanism to preserve historical elite solutions, alpha wolf tenure inspection and rotation to maintain population vitality, and a hybrid mutation strategy combining differential evolution and Cauchy perturbations to enhance diversity and global search capability. Ablation studies validate the efficacy of each strategy, while simulation experiments demonstrate FMGWO’s superior performance in WSN coverage optimization. Compared to established algorithms such as PSO, GWO, CSA, DE, GA, FA, OGWO, DGWO1, and DGWO2, FMGWO achieves higher coverage rates with fewer nodes—up to 98.63% with 30 nodes—alongside improved convergence speed and stability. These results underscore FMGWO’s potential as an effective solution for efficient WSN deployment, offering significant implications for resource-constrained optimization in IoT and edge computing systems. Full article

To address the triple bottlenecks of data scarcity, oversized models, and slow inference that hinder Cantonese automatic speech recognition (ASR) in low-resource and edge-deployment settings, this study proposes a cost-effective Cantonese ASR system based on LoRA fine-tuning and INT8 quantization. First, Whisper-tiny is parameter-efficiently fine-tuned on the Common Voice zh-HK training set using LoRA with rank = 8. Only 1.6% of the original weights are updated, reducing the character error rate (CER) from 49.5% to 11.1%, a performance close to full fine-tuning (10.3%), while cutting the training memory footprint and computational cost by approximately one order of magnitude. Next, the fine-tuned model is compressed into a 60 MB INT8 checkpoint via dynamic quantization in ONNX Runtime. On a MacBook Pro M1 Max CPU, the quantized model achieves an RTF = 0.20 (offline inference 5 × real-time) and 43% lower latency than the FP16 baseline; on an NVIDIA A10 GPU, it reaches RTF = 0.06, meeting the requirements of high-concurrency cloud services. Ablation studies confirm that the LoRA-INT8 configuration offers the best trade-off among accuracy, speed, and model size. Limitations include the absence of spontaneous-speech noise data, extreme-hardware validation, and adaptive LoRA structure optimization. Future work will incorporate large-scale self-supervised pre-training, tone-aware loss functions, AdaLoRA architecture search, and INT4/NPU quantization, and will establish an mJ/char energy–accuracy curve. The ultimate goal is to achieve CER ≤ 8%, RTF < 0.1, and mJ/char < 1 for low-power real-time Cantonese ASR in practical IoT scenarios. Full article

Grassland aboveground biomass (AGB) serves as a critical indicator of ecosystem productivity and carbon cycling, playing a pivotal role in ecosystem functioning. The advances in hyperspectral and terrestrial Light Detection and Ranging (LiDAR) data have provided new opportunities for grassland AGB monitoring, but current research remains predominantly focused on data-driven machine learning models. The black-box nature of such approaches resulted in a lack of clear interpretation regarding the coupling relationships between these two data types in grassland AGB estimation. For grassland aboveground fresh biomass, the theoretical estimation can be decomposed into either the product of planar density (PD) and plot area or the product of stereoscopic density (SD) and grassland community volume. Based on this theory, our study developed a semi-mechanistic remote sensing model for grassland AGB estimation by integrating hyperspectral-derived biomass density with extracted structural parameters from terrestrial LiDAR. Initially, we built hyperspectral estimation models for both PD and SD of grassland fresh AGB using PLSR. Subsequently, by integrating the inversion results with grassland quadrat area and community volume measurements, respectively, we achieved quadrat-scale remote sensing estimation of grassland AGB. Finally, we conducted comparative accuracy assessments of both methods across different phenological stages to evaluate their performance differences. Our results demonstrated that SD, which incorporated structural features, could be more precisely estimated (R² = 0.90, nRMSE = 7.92%, Bias% = 0.01%) based on hyperspectral data compared to PD (R² = 0.79, nRMSE = 10.19%, Bias% = −7.25%), with significant differences observed in their respective responsive spectral bands. PD showed greater sensitivity to shortwave infrared regions, while SD exhibited stronger associations with visible, red-edge, and near-infrared bands. Although both methods achieved comparable overall AGB estimation accuracy (PD-based: R² = 0.79, nRMSE = 10.19%, Bias% = −7.25%; SD-based: R² = 0.82, nRMSE = 10.58%, Bias% = 1.86%), the SD-based approach effectively mitigated the underestimation of high biomass values caused by spectral saturation effects and also demonstrated superior and more stable performance across different growth periods (R² > 0.6). This work provided concrete physical meaning to the integration of hyperspectral and LiDAR data for grassland AGB monitoring and further suggested the potential of multi-source remote sensing data fusion in estimating grassland AGB. The findings offered theoretical foundations for developing large-scale grassland AGB monitoring models using airborne and spaceborne remote sensing platforms. Full article

All-solid-state lithium batteries (ASSLBs) are emerging as a promising alternative to conventional lithium-ion batteries, offering solutions to challenges related to energy density and safety. Their core advancement relies on breakthroughs in solid-state electrolytes (SEs). SEs can be broadly grouped into two main types: inorganic solid electrolytes (ISEs) and organic solid electrolytes (OSEs). ISEs offer high ionic conductivity (0.1~1 mS cm⁻¹), a lithium-ion transference number close to 1, and excellent thermal stability, but their intrinsic brittleness leads to poor interfacial wettability and processing difficulties, limiting practical applications. In contrast, OSEs exhibit good flexibility and interfacial compatibility but suffer from poor ionic conductivity (10⁻⁴~10⁻² mS cm⁻¹) due to high crystallinity at room temperature, in addition to poor thermal stability and weak mechanical integrity, making it difficult to match high-voltage cathodes and suppress lithium dendrite growth. Against this backdrop, the stability of the organic–inorganic interface plays a crucial role. However, challenges such as low overall conductivity and unstable interfaces still limit their performance. This review provides a microscopic perspective on lithium-ion transport pathways across the polymer phase, the inorganic filler phase, and their interfacial regions. It categorizes inert fillers and active fillers, analyzing their structure–performance relationships and emphasizing the synergistic effects of filler dimensionality, surface chemistry, and interfacial interactions. In addition, cutting-edge analytical methods such as time-of-flight secondary ion mass spectrometry (TOF-SIMS) and high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) have also been employed and are summarized into their roles for revealing the microstructures and dynamic interfacial behaviors of OICSEs. Finally, future directions are proposed, such as hierarchical pore structure design, surface functionalization, and simulation-guided optimization, aiming to provide theoretical insights and technological strategies for the development of high-performance composite electrolytes for ASSLBs. Full article

With the growing adoption of circular economy principles in manufacturing, efficient disassembly and reassembly of end-of-life (EOL) products has become a key challenge in smart factories. This paper addresses the Disassembly and Assembly Line Balancing Problem (DALBP), which involves scheduling robotic tasks across workstations while minimizing total operation time and accounting for directional switching time between disassembly and assembly phases. To solve this problem, we propose an improved reinforcement learning algorithm, IQ($\lambda$), which extends the classical Q($\lambda$) method by incorporating eligibility trace decay, a dynamic Action Table mechanism to handle non-conflicting parallel tasks, and switching-aware reward shaping to penalize inefficient task transitions. Compared with standard Q($\lambda$), these modifications enhance the algorithm’s global search capability, accelerate convergence, and improve solution quality in complex DALBP scenarios. While the current implementation does not deploy live IoT infrastructure, the architecture is modular and designed to support future extensions involving edge-cloud coordination, trust-aware optimization, and privacy-preserving learning in Industrial Internet of Things (IIoT) environments. Four real-world disassembly-assembly cases (flashlight, copier, battery, and hammer drill) are used to evaluate the algorithm’s effectiveness. Experimental results show that IQ($\lambda$) consistently outperforms traditional Q-learning, Q($\lambda$), and Sarsa in terms of solution quality, convergence speed, and robustness. Furthermore, ablation studies and sensitivity analysis confirm the importance of the algorithm’s core design components. This work provides a scalable and extensible framework for intelligent scheduling in cyber-physical manufacturing systems and lays a foundation for future integration with secure, IoT-connected environments. Full article