Search Results (549)

Deep coalbed methane (CBM) reservoirs are characterized by high hydrocarbon content and are considered an important strategic resource. Due to their inherently low permeability and porosity, horizontal well drilling is commonly employed to enhance production, with the length of the horizontal section playing a critical role in determining CBM output. However, during extended horizontal drilling, wellbore instability frequently occurs as a result of drilling fluid invasion into the coal formation, posing significant safety challenges. This instability is primarily caused by the physical intrusion of drilling fluids and their interactions with the coal seam, which alter the mechanical integrity of the formation. To address these challenges, interpenetrating and semi-interpenetrating network (IPN/s-IPN) hydrogels have gained attention due to their superior physicochemical properties. This material offers enhanced sealing and support performance across fracture widths ranging from micrometers to millimeters, making it especially suited for plugging applications in deep CBM reservoirs. A self-degradable interpenetrating double-network hydrogel particle plugging agent (SSG) was developed in this study, using polyacrylamide (PAM) as the primary network and an ionic polymer as the secondary network. The SSG demonstrated excellent thermal stability, remaining intact for at least 40 h in simulated formation water at 120 °C with a degradation rate as high as 90.8%, thereby minimizing potential damage to the reservoir. After thermal aging at 120 °C, the SSG maintained strong plugging performance and favorable viscoelastic properties. A drilling fluid containing 2% SSG achieved an invasion depth of only 2.85 cm in an 80–100 mesh sand bed. The linear viscoelastic region (LVR) ranged from 0.1% to 0.98%, and the elastic modulus reached 2100 Pa, indicating robust mechanical support and deformation resistance. Full article

The global cement industry remains a significant contributor to carbon dioxide (CO₂) emissions, prompting substantial research efforts toward sustainable construction materials. Lithium slag (LS), a by-product of lithium extraction, has attracted attention as a supplementary cementitious material (SCM). This review synthesizes experimental findings on LS replacement levels, fresh-state behavior, mechanical performance (compressive, tensile, and flexural strengths), time-dependent deformation (shrinkage and creep), and durability (sulfate, acid, abrasion, and thermal) of LS-modified concretes. Statistical analysis identifies an optimal LS dosage of 20–30% (average 24%) for maximizing compressive strength and long-term durability, with 40% as a practical upper limit for tensile and flexural performance. Fresh-state tests show that workability losses at high LS content can be mitigated via superplasticizers. Drying shrinkage and creep strains decrease in a dose-dependent manner with up to 30% LS. High-volume (40%) LS blends achieve up to an 18% gain in 180-day compressive strength and >30% reduction in permeability metrics. Under elevated temperatures, 20% LS mixes retain up to 50% more residual strength than controls. In advanced systems—autoclaved aerated concrete (AAC), one-part geopolymers, and recycled aggregate composites—LS further enhances both microstructural densification and durability. In particular, LS emerges as a versatile SCM that optimizes mechanical and durability performance, supports material circularity, and reduces the carbon footprint. Full article

To overcome the limitations of traditional image alignment methods in capturing deep semantic features, a deep feature information image alignment network (DFA-Net) is proposed. This network aims to enhance image alignment performance through multi-level feature learning. DFA-Net is based on the deep residual architecture and introduces spatial pyramid pooling to achieve cross-scalar feature fusion, effectively enhancing the feature’s adaptability to scale. A feature enhancement module based on the self-attention mechanism is designed, with key features that exhibit geometric invariance and high discriminative power, achieved through a dynamic weight allocation strategy. This improves the network’s robustness to multimodal image deformation. Experiments on two public datasets, MSRS and RoadScene, show that the method performs well in terms of alignment accuracy, with the RMSE metrics being reduced by 0.661 and 0.473, and the SSIM, MI, and NCC improved by 0.155, 0.163, and 0.211; and 0.108, 0.226, and 0.114, respectively, compared with the benchmark model. The visualization results validate the significant improvement in the features’ visual quality and confirm the method’s advantages in terms of stability and discriminative properties of deep feature extraction. Full article

Foreign Object Debris (FOD) detection is paramount for airport operations. The precise identification and removal of FOD are critical for ensuring airplane flight safety. This study collected FOD images using optical imaging sensors installed at Urumqi Airport and created a custom FOD dataset based on these images. To address the challenges of small targets and complex backgrounds in the dataset, this paper proposes optimizations and improvements based on the advanced detection network Dab-Deformable. First, this paper introduces a Lightweight Deep-Shallow Feature Fusion algorithm (LDSFF), which integrates a hotspot sensing network and a spatial mapping enhancer aimed at focusing the model on significant regions. Second, we devise a Multi-Directional Deformable Channel Attention (MDDCA) module for rational feature weight allocation. Furthermore, a feedback mechanism is incorporated into the encoder structure, enhancing the model’s capacity to capture complex dependencies within sequential data. Additionally, when combined with a Threshold Selection (TS) algorithm, the model effectively mitigates the distraction caused by the serialization of multi-layer feature maps in the Transformer architecture. Experimental results on the optical small FOD dataset show that the proposed network achieves a robust performance and improved accuracy in FOD detection. Full article

3D object detection is a fundamental task in autonomous driving. In recent years, voxel-based methods have demonstrated significant advantages in reducing computational complexity and memory consumption when processing large-scale point cloud data. A representative method, Voxel-RCNN, introduces Region of Interest (RoI) pooling on voxel features, successfully bridging the gap between voxel and point cloud representations for enhanced 3D object detection. However, its robustness deteriorates when detecting distant objects or in the presence of noisy points (e.g., traffic signs and trees). To address this limitation, we propose an enhanced approach named Self-Attention Voxel-RCNN (SA-VoxelRCNN). Our method integrates two complementary attention mechanisms into the feature extraction phase. First, a full self-attention (FSA) module improves global context modeling across all voxel features. Second, a deformable self-attention (DSA) module enables adaptive sampling of representative feature subsets at strategically selected positions. After extracting contextual features through attention mechanisms, these features are fused with spatial features from the base algorithm to form enhanced feature representations, which are subsequently input into the region proposal network (RPN) to generate high-quality 3D bounding boxes. Experimental results on the KITTI test set demonstrate that SA-VoxelRCNN achieves consistent improvements in challenging scenarios, with gains of 2.49 and 1.87 percentage points at Moderate and Hard difficulty levels, respectively, while maintaining real-time performance at 22.3 FPS. This approach effectively balances local geometric details with global contextual information, providing a robust detection solution for autonomous driving applications. Full article

Camouflaged Object Detection (COD) aims to segment objects that are highly integrated with their background, presenting significant challenges such as low contrast, complex textures, and blurred boundaries. Existing deep learning methods often struggle to achieve robust segmentation under these conditions. To address these limitations, this paper proposes a novel COD network, SAM2-DFBCNet, built upon the SAM2 Hiera architecture. Our network incorporates three key modules: (1) the Camouflage-Aware Context Enhancement Module (CACEM), which fuses local and global features through an attention mechanism to enhance contextual awareness in low-contrast scenes; (2) the Cross-Scale Feature Interaction Bridge (CSFIB), which employs a bidirectional convolutional GRU for the dynamic fusion of multi-scale features, effectively mitigating representation inconsistencies caused by complex textures and deformations; and (3) the Dynamic Boundary Refinement Module (DBRM), which combines channel and spatial attention mechanisms to optimize boundary localization accuracy and enhance segmentation details. Extensive experiments on three public datasets—CAMO, COD10K, and NC4K—demonstrate that SAM2-DFBCNet outperforms twenty state-of-the-art methods, achieving maximum improvements of 7.4%, 5.78%, and 4.78% in key metrics such as S-measure ($S_{\alpha }$), F-measure ($F_{\beta }$), and mean E-measure ($E_{\varphi }$), respectively, while reducing the Mean Absolute Error (M) by 37.8%. These results validate the superior performance and robustness of our approach in complex camouflage scenarios. Full article

Three-dimensional (3D) concrete printing (3DCP) has gained significant attention over the last decade due to its many claimed benefits. The absence of effective real-time quality control mechanisms, however, can lead to inconsistencies in extrusion, compromising the integrity of 3D-printed structures. Although the importance of quality control in 3DCP is broadly acknowledged, research lacks systematic methods. This research investigates the feasibility of using ultrasonic pulse velocity (UPV) as a practical, in situ, real-time monitoring tool for 3DCP. Two different groups of binders were investigated: limestone calcined clay (LC³) and zeolite-based mixes in binary and ternary blends. Filaments of 200 mm were extruded every 5 min, and UPV, pocket hand vane, flow table, and viscometer tests were performed to measure pulse velocity, shear strength, relative deformation, yield stress, and plastic viscosity, respectively, in the fresh state. Once the filaments presented printing defects (e.g., filament tearing, filament width reduction), the tests were concluded, and the open time was recorded. Isothermal calorimetry tests were conducted to obtain the initial heat release and reactivity of the supplementary cementitious materials (SCMs). Results showed a strong correlation (R² = 0.93) between UPV and initial heat release, indicating that early hydration (ettringite formation) influenced UPV and determined printability across different mixes. No correlation was observed between the other tests and hydration kinetics. UPV demonstrated potential as a real-time monitoring tool, provided the mix-specific pulse velocity is established beforehand. Further research is needed to evaluate UPV performance during active printing when there is an active flow through the printer. Full article

Biomimetic lattice structures, inspired by natural architectures such as bone, coral, mollusk shells, and Euplectella aspergillum, have gained increasing attention for their exceptional strength-to-weight ratios, energy absorption, and deformation control. These properties make them ideal for advanced engineering applications in aerospace, biomedical devices, and structural impact protection. This study presents a comprehensive bibliometric analysis of global research on biomimetic lattice structures published between 2020 and 2025, aiming to identify thematic trends, collaboration patterns, and underexplored areas. A curated dataset of 3685 publications was extracted from databases like PubMed, Dimensions, Scopus, IEEE, Google Scholar, and Science Direct and merged together. After the removal of duplication and cleaning, about 2226 full research articles selected for the bibliometric analysis excluding review works, conference papers, book chapters, and notes using Cite space, VOS viewer version 1.6.20, and Bibliometrix R packages (4.5. 64-bit) for mapping co-authorship networks, institutional affiliations, keyword co-occurrence, and citation relationships. A significant increase in the number of publications was found over the past year, reflecting growing interest in this area. The results identify China as the most prolific contributor, with substantial institutional support and active collaboration networks, especially with European research groups. Key research focuses include additive manufacturing, finite element modeling, machine learning-based design optimization, and the performance evaluation of bioinspired geometries. Notably, the integration of artificial intelligence into structural modeling is accelerating a shift toward data-driven design frameworks. However, gaps remain in geometric modeling standardization, fatigue behavior analysis, and the real-world validation of lattice structures under complex loading conditions. This study provides a strategic overview of current research directions and offers guidance for future interdisciplinary exploration. The insights are intended to support researchers and practitioners in advancing next-generation biomimetic materials with superior mechanical performance and application-specific adaptability. Full article

Liquid metals (LMs), with their unique combination of high electrical conductivity and mechanical deformability, have emerged as promising materials for stretchable electronics and biointerfaces. However, the practical application of bulk LMs in wearable sensors has been hindered by processing challenges and low stability. To overcome these limitations, liquid metal particles (LMPs) encapsulated by native oxide shells have gained attention as versatile and stable fillers for stretchable and conductive composites. Recent advances have focused on the development of LM-based hybrid composites that combine LMPs with metal, carbon, or polymeric fillers. These systems offer enhanced electrical and mechanical properties and can form conductive networks without the need for additional sintering processes. They also impart composites with multiple functions such as self-healing, electromagnetic interference shielding, and recyclability. Hence, the present review summarizes the fabrication methods and functional properties of LM-based composites, with a particular focus on their applications in wearable sensing. In addition, recent developments in the use of LM composites for physical motion monitoring (e.g., strain and pressure sensing) and electrophysiological signal recording (e.g., EMG and ECG) are presented, and the key challenges and opportunities for next-generation wearable platforms are discussed. Full article

With the exponential growth of engineering monitoring data, data-driven neural networks have gained widespread application in predicting retaining structure deformation in foundation pit engineering. However, existing models often overlook the spatial deflection correlations among monitoring points. Therefore, this study proposes a novel deep learning framework, CGCA (Convolutional Gated Recurrent Unit with Cross-Attention), which integrates ConvGRU and cross-attention mechanisms. The model achieves spatio-temporal feature extraction and deformation prediction via an encoder–decoder architecture. Specifically, the convolutional structure captures spatial dependencies between monitoring points, while the recurrent unit extracts time-series characteristics of deformation. A cross-attention mechanism is introduced to dynamically weight the interactions between spatial and temporal data. Additionally, the model incorporates multi-dimensional inputs, including full-depth inclinometer measurements, construction parameters, and tube burial depths. The optimization strategy combines AdamW and Lookahead to enhance training stability and generalization capability in geotechnical engineering scenarios. Case studies of foundation pit engineering demonstrate that the CGCA model exhibits superior performance and robust generalization capabilities. When validated against standard section (CX1) and complex working condition (CX2) datasets involving adjacent bridge structures, the model achieves determination coefficients (R²) of 0.996 and 0.994, respectively. The root mean square error (RMSE) remains below 0.44 mm, while the mean absolute error (MAE) is less than 0.36 mm. Comparative experiments confirm the effectiveness of the proposed model architecture and the optimization strategy. This framework offers an efficient and reliable technical solution for deformation early warning and dynamic decision-making in foundation pit engineering. Full article

Sign language is a complex and dynamic visual language that requires the coordinated movement of various body parts, such as the hands, arms, and limbs—making it an ideal application domain for sensor networks to capture and interpret human gestures accurately. To address the intricate task of precise and expedient SLR from raw videos, this study introduces a novel deep learning approach by devising a multimodal framework for SLR. Specifically, feature extraction models are built based on two modalities: skeleton and RGB images. In this paper, we firstly propose a Multi-Stream Spatio-Temporal Graph Convolutional Network (MSGCN) that relies on three modules: a decoupling graph convolutional network, a self-emphasizing temporal convolutional network, and a spatio-temporal joint attention module. These modules are combined to capture the spatio-temporal information in multi-stream skeleton features. Secondly, we propose a 3D ResNet model based on deformable convolution (D-ResNet) to model complex spatial and temporal sequences in the original raw images. Finally, a gating mechanism-based Multi-Stream Fusion Module (MFM) is employed to merge the results of the two modalities. Extensive experiments are conducted on the public datasets AUTSL and WLASL, achieving competitive results compared to state-of-the-art systems. Full article

Atmospheric turbulence severely degrades high-resolution satellite videos through spatiotemporally coupled distortions, including temporal jitter, spatial-variant blur, deformation, and scintillation, thereby constraining downstream analytical capabilities. Restoring turbulence-corrupted videos poses a challenging ill-posed inverse problem due to the inherent randomness of turbulent fluctuations. While existing turbulence mitigation methods for long-range imaging demonstrate partial success, they exhibit limited generalizability and interpretability in large-scale satellite scenarios. Inspired by refractive-index structure constant ($C_n^2$) estimation from degraded sequences, we propose a physics-informed turbulence signature (TS) prior that explicitly captures spatiotemporal distortion patterns to enhance model transparency. Integrating this prior into a lucky imaging framework, we develop a Physics-Based Turbulence Mitigation Network guided by Turbulence Signature (TMTS) to disentangle atmospheric disturbances from satellite videos. The framework employs deformable attention modules guided by turbulence signatures to correct geometric distortions, iterative gated mechanisms for temporal alignment stability, and adaptive multi-frame aggregation to address spatially varying blur. Comprehensive experiments on synthetic and real-world turbulence-degraded satellite videos demonstrate TMTS’s superiority, achieving 0.27 dB PSNR and 0.0015 SSIM improvements over the DATUM baseline while maintaining practical computational efficiency. By bridging turbulence physics with deep learning, our approach provides both performance enhancements and interpretable restoration mechanisms, offering a viable solution for operational satellite video processing under atmospheric disturbances. Full article

Flexible inflatable film wings have many functional advantages that traditional fixed rigid wings do not possess, such as foldability, small size, light weight, convenient storage, transportation, and so on. More and more scholars and engineers are paying attention to flexible inflatable wings, which have gradually become a new hot research topic. However, flexible wings rely on inflation pressure to maintain the shape and rigidity of the skin film, and the inflation pressure has a significant influence on the strain deformation and wing bearing characteristics of flexible wing skin film. Here, based on the flexible mechanics theory and balance principle of flexible inflatable film, a theoretical model of structural deformation and internal inflation pressure was constructed, and finite element simulation analysis under different internal inflation pressure conditions was carried out as well. The results demonstrate that the biaxial deformation of flexible wing skin film is closely related to internal inflation pressure, local size, configuration, and film material properties. However, strain deformation along the wingspan direction is quite distinguishing, skin films work under the condition of biaxial plane deformation, and the strain deformation of the spanning direction is obviously higher than that of the chord direction, which all increases with internal inflation pressure. Therefore, it is necessary to pay more attention to bearing strain deformation characteristics to meet the bearing stiffness requirements, which could effectively provide a theoretical reference for the structural optimization design and inflation scheme setting of flexible inflatable wings. Full article

The intricate and complex tumor cell morphology in breast pathology images is a key factor for tumor classification. This paper proposes a lightweight breast tumor classification model with multi-frequency feature fusion (LMFM) to tackle the problem of inadequate feature extraction and poor classification performance. The LMFM utilizes wavelet transform (WT) for multi-frequency feature fusion, integrating high-frequency (HF) tumor details with high-level semantic features to enhance feature representation. The network’s ability to extract irregular tumor characteristics is further reinforced by dynamic adaptive deformable convolution (DADC). The introduction of the token-based Region Focus Module (TRFM) reduces interference from irrelevant background information. At the same time, the incorporation of a linear attention (LA) mechanism lowers the model’s computational complexity and further enhances its global feature extraction capability. The experimental results demonstrate that the proposed model achieves classification accuracies of 98.23% and 97.81% on the BreaKHis and BACH datasets, with only 9.66 M parameters. Full article

With the rapid development of wireless communication technologies and the increasing demand for efficient spectrum utilization, broadband spectrum sensing has become critical in both civilian and military fields. Detecting narrowband signals under broadband environments, especially under low-signal-to-noise-ratio (SNR) conditions, poses significant challenges due to the complexity of time–frequency features and noise interference. To this end, this study presents a signal detection model named deformable feature-enhanced network–You Only Look Once (DFN-YOLO), specifically designed for blind signal detection in broadband scenarios. The DFN-YOLO model incorporates a deformable channel feature fusion network (DCFFN), replacing the concatenate-to-fusion (C2f) module to enhance the extraction and integration of channel features. The deformable attention mechanism embedded in DCFFN adaptively focuses on critical signal regions, while the loss function is optimized to the focal scaled intersection over union (Focal_SIoU), improving detection accuracy under low-SNR conditions. To support this task, a signal detection dataset is constructed and utilized to evaluate the performance of DFN-YOLO. The experimental results for broadband time–frequency spectrograms demonstrate that DFN-YOLO achieves a mean average precision (mAP50–95) of 0.850, averaged over IoU thresholds ranging from 0.50 to 0.95 with a step of 0.05, significantly outperforming mainstream object detection models such as YOLOv8, which serves as the benchmark baseline in this study. Additionally, the model maintains an average time estimation error within $5.55\times {10}^{-5}$ s and provides preliminary center frequency estimation in the broadband spectrum. These findings underscore the strong potential of DFN-YOLO for blind signal detection in broadband environments, with significant implications for both civilian and military applications. Full article