Search Results (513)

Accurate multiphase flow measurement in horizontal wells is fundamentally challenged by the antagonistic electrical responses of water and gas: Electrical Resistance Tomography (ERT) loses sensitivity to thin liquid films, while Electrical Capacitance Tomography (ECT) suffers signal saturation in conductive water, preventing either modality from covering the full operating envelope alone. This study proposes a physics-guided hybrid modeling framework that integrates multi-modal ERT/ECT sensing to achieve high-precision flowrate inversion. The framework utilizes a corrected multi-modal fusion algorithm, achieving a liquid holdup MAPE of 2.5 ± 0.5% representing a nearly two-fold improvement over the best single-modality system (Direct ERT, 4.5%). For velocity estimation, an optimized cross-correlation method yields results with ± 3.0% error, incorporating multi-sensor and multi-sequence fusion. A key finding is that deep neural networks exhibit Architectural Phase Specialization: multi-branch architectures (MB-DNN) perform strongly on localized, heterogeneous liquid structures (2.0% liquid error), whereas fully-connected architectures (FC-DNN) excel at capturing the global patterns of the continuous gas core (1.2% gas error). By hybridizing a calibrated drift-flux physical model with these phase-specialized DNNs, the framework achieves overall averaged errors of 1.8% for gas and 1.5% for liquid across the full experimental envelope. The proposed framework was evaluated on 444,313 experimental samples and subsequently validated in a three-month industrial trial at the Puguang gas field under extreme conditions (26 MPa, 80 °C), where it maintained a prediction error of ± 2.3%. This work establishes a scalable, physically consistent paradigm for intelligent hydrocarbon production monitoring. Full article

Mulching is an agronomic practice that improves orchard soil and promotes root growth. To investigate the regulatory effects of different mulching materials on soil properties, microbial communities, and root function in apple orchards, eight treatments were established: clean tillage (CK), organic fertilizer mulching (OFM), chopped corn straw mulching (SM1), chopped and bundled corn straw mulching (SM2), intact corn stover mulching (SM3), composted apple branch mulching (BM), horticultural ground cover fabric mulching (FM), and weed mulching (WM). The results showed that OFM, BM, SM1, and SM3 exhibited effective cooling effects during summer. During the peak root-flush period, OFM, SM3, and BM significantly reduced soil bulk density, increased porosity, enhanced soil organic matter and available nutrient contents, and elevated the activities of soil sucrase, urease, and catalase. Moreover, these treatments promoted the accumulation of carbohydrates and the uptake of mineral nutrients in roots. OFM and SM3 significantly increased the Simpson index of both soil bacterial and fungal communities, while BM improved the beta diversity of bacterial and fungal communities. OFM, SM3, and BM can effectively improve soil physicochemical properties, optimize microbial community structure, and enhance root nutrient uptake. It is recommended as a mulching measure for soil in northern apple orchards. Among the eight treatments evaluated, OFM, SM3, and BM exhibited superior performance in improving soil physicochemical properties, promoting root function, and enhancing microbial community diversity. Therefore, the findings of this study provide an effective soil management strategy for apple orchards in the cold northern regions of China. Full article

Beef cattle behavior provides valuable information regarding their health status. Recently, deep convolutional network-based methods have achieved considerable results in beef cattle behavior recognition. However, their robustness under low-light or dark conditions remains limited, which restricts their application in real farm environments. To address this issue, this study constructed a realistic beef cattle behavior dataset in the dark, named Dark Beef Cattle Actions, which was collected under real nighttime farm conditions. The constructed dataset contains 1097 video clips collected from 30 beef cattle and covers 6 behavioral classes, including running, feeding, drinking, grooming, mounting, and fighting. Based on this dataset, we proposed a novel neural network architecture based on spatio-temporal dark enhancement and dynamic fusion for beef cattle behavior recognition in the dark. First, a spatio-temporal dark enhancement module was designed to improve dark video quality while preserving motion features. Second, a dynamic fusion module was introduced to adaptively fuse features from different branches and obtain more discriminative representations. In addition, a joint loss was adopted to optimize both dark enhancement and action recognition. Experimental results on the constructed dataset show that the proposed method achieved a weighted-averaged precision score of 88.47%, a weighted-averaged recall score of 80.18%, an accuracy score of 83.80%, and a weighted-averaged F1-score of 84.12%. Compared with other state-of-the-art methods, the proposed method achieved competitive performance in the recognition of night-time beef cattle behavior. These findings would provide support for intelligent livestock behavior recognition and monitoring in precision farming. Full article

Recent advances in multi-modal large language models have enabled systems to jointly process text, images, and audio. However, these developments have primarily benefited high-resource languages, leaving many low-resource communities underserved. In response, we introduce Qolda-AVL, a compact five-billion-parameter audio–vision–language model tailored for Kazakh. Qolda-AVL extends our previous Qolda vision–language model by adding a dedicated audio perception branch while maintaining strong visual and linguistic performance. Built on the Qwen3-VL-Thinking backbone, we incorporate Audio DeepStack, which transfers features from three intermediate Whisper encoder layers into the first three layers of the language model using dedicated projections and residual connections. The model is trained through a four-stage pipeline: adapting the Whisper encoder and language model to Kazakh, aligning the new audio branch to the language backbone, and jointly fine-tuning all modules on chain-of-thought reasoning tasks across audio, image, and text. All audio, vision, and language capabilities are evaluated using the model’s native reasoning mode, and a chain-of-thought trace is generated before each final answer during the performance assessment. To facilitate further research, we open-source the model along with the adapted Kazakh versions of four audio benchmarks, covering spoken attribute reasoning, spoken mathematical question answering, and audio captioning with question answering. Full article

Joint classification of hyperspectral images (HSI) and light detection and ranging (LiDAR) data is important for land-cover recognition, as it can exploit both spectral discrimination and structural elevation information. However, existing methods mainly focus on visual feature fusion and insufficiently utilize class-level semantic priors, which limits their discriminative capability in complex boundaries, visually similar categories, and limited-sample scenarios. To address these issues, this paper proposes a text-guided multimodal semantic fusion network for HSI–LiDAR classification. Specifically, a Channel-Modulated Mobile Convolution Module (CMMC) is designed to extract modality-specific features, a Spatial–Frequency Feature Enhancement Module (SFFE) is introduced to enhance spatial-boundary and frequency-domain structural representations, and a Bidirectional Cross-Modal Fusion Module (BCMF) is developed to promote complementary interaction between spectral and structural information. Meanwhile, class-level textual descriptions are constructed from class names, color attributes, and geographical contexts, and a text encoder is employed to obtain semantic prototypes. Furthermore, a multi-branch vision–text semantic alignment mechanism projects HSI features, LiDAR features, and fused features into a shared semantic space for joint constraints, improving semantic consistency and class separability. Experiments on the Houston2013, Augsburg, and Trento datasets demonstrate the effectiveness of the proposed method. It achieves an overall accuracy of 98.76% on Houston2013, with improvements of 0.62%, 0.52%, and 0.67 in overall accuracy, average accuracy, and Kappa coefficient × 100 over the best competing results, respectively. The proposed method also obtains the best overall metrics on Augsburg and Trento, and ablation studies verify the effectiveness of the proposed components. Full article

Existing research on the axial compressive performance and size effect of carbon fiber-reinforced polymer (CFRP)-confined recycled aggregate concrete (RAC) short columns mainly relies on macroscopic experimental analysis, lacking research methods capable of reflecting the heterogeneous characteristics of materials and mesoscopic damage evolution mechanisms. Accordingly, a three-dimensional mesoscale finite element method was adopted in this study to establish a five-phase RAC mesoscopic model, including natural aggregates, old mortar, old interfacial transition zones (ITZs), new mortar, and new interfacial transition zones. Different from existing studies, predominantly based on macroscopic experiments or empirical models, this paper focuses on revealing the coupled effects of the recycled aggregate replacement ratio, the number of CFRP confinement layers, and specimen size. A total of 48 specimens were designed, covering four specimen sizes, four recycled coarse aggregate replacement ratios, and three CFRP confinement layers. The effects of these parameters on failure modes, stress–strain relationships, and size effect were systematically analyzed. The results indicate that the peak stress decreases significantly with the increase in the recycled coarse aggregate replacement ratio; the increase in CFRP layers markedly improves both the bearing capacity and post-peak bearing capacity retention rate; the ultimate stress generally declines as the specimen size increases, which highlights the pronounced size effect of CFRP-confined RAC short columns. Based on peak parameters and normalization analysis, a simplified stress–strain model was established: the goodness of fit R² of the ascending branch is 0.98565, and the goodness of fit for the descending branch parameters are ${\mathrm{R}}_{\mathsf{\beta }}^2$= 0.9655 and ${\mathrm{R}}_{\mathsf{\gamma }}^2$= 0.9350. Compared with existing models, the proposed model achieves a low prediction error of only 1.5–6.9%, demonstrating superior prediction accuracy. It can accurately describe the complete compressive process of CFRP-confined RAC short columns and provide a mesoscopic mechanistic basis for engineering design. Full article

Accurate lake boundary extraction from optical remote sensing imagery remains challenging in high-altitude regions such as the Tibetan Plateau due to ice cover, snow, shadows, and spectrally similar backgrounds. Although recent deep learning models achieve strong region-overlap performance, they often fail to ensure stable shoreline localization. To address this issue, we propose a Global–Local Residual Refinement Network (GLR-Net) for boundary-aware lake extraction from remote sensing imagery. The proposed framework first captures large-scale semantic context through a global branch and subsequently performs patch-level residual refinement to improve local shoreline geometry. A global-to-local guidance mechanism is further introduced to incorporate structural priors into local refinement. Experiments on a manually annotated Tibetan Plateau lake dataset demonstrate that the proposed method achieves competitive region-level segmentation performance while improving geometric shoreline accuracy. Compared with representative semantic segmentation baselines, including U-Net, SegFormer-B0, SegFormer-B4, and OCRNet, the proposed method achieves the highest Boundary F1 score of 0.811 under a 3-pixel tolerance and the lowest mean BDE of 13.19 pixels. The results indicate that conventional overlap-based metrics alone are insufficient for evaluating shoreline delineation quality in complex alpine environments. Full article

Controllable symbolic music generation must preserve a reference melody while remaining responsive to style prompts. Existing hierarchical diffusion systems typically reuse a shared condition vector across harmony, rhythm, and timbre stages, which can entangle stylistic factors and weaken melody preservation. We present HCDMG++, a hierarchical diffusion framework that addresses these two limitations through stage-aware style routing and differentiable melody regularization. The routing module uses a residual multi-layer perceptron (MLP) with zero-initialized scalar gates to project text-derived style embeddings into harmony-, rhythm-, and timbre-specific subspaces, whereas the regularization branch aligns soft pitch histograms and contour trajectories with the conditioning melody during training without breaking the differentiable computation graph. We evaluate the integrated system on a 384-sample benchmark covering four melodies, eight styles, four random seeds, and three denoising budgets, supplemented by a matched legacy-compatible reference and inference-time component ablation that contrasts legacy behavior, silenced gates, an automated uniform gamma routing sweep, and the full forward pass. HCDMG++ produces valid four-track outputs in all 384 runs, reaches a peak pitch histogram similarity score of 0.508 under a 64-step budget, and improves pitch histogram alignment over Legacy-HCDMG by roughly two orders of magnitude on the matched slice, while attaining a positive Fisher-style style separability score where the legacy benchmark is too sparse to support one. These results indicate that stage-specific conditioning and differentiable structural guidance jointly improve controllability in symbolic music diffusion, while also exposing the remaining limitations in long-form generalization and perceptual validation, which motivate the future work outlined at the end of this paper. Full article

Hyperspectral image (HSI) classification enables precise classification of land-cover types from rich spectral and spatial information. Recent methods combine convolutional neural network (CNN) and Mamba branches to exploit their complementary local and global modeling capabilities for HSI classification. However, most of these methods treat all spectral channels uniformly in feature fusion, failing to account for the discriminability differences across spectral bands. Moreover, most methods rely on a single classification head at the final layer, which may lead to vanishing gradients in shallow layers. To address these limitations, a spectral group-wise gated CNN–Mamba network with cross-stage mutual distillation, called SGGCMNet, is proposed. To address the first limitation, a CNN–Mamba spectral group-wise gating block (CMSB) is designed at the feature-fusion level. Specifically, the CMSB partitions channels into multiple sub-groups along the spectral dimension. Each sub-group learns its own fusion weights that balance local spectral–spatial cues produced by a CNN pathway with long-range context produced by a Mamba pathway. To address the second limitation, two loss-level optimization strategies are proposed jointly: A progressive deep supervision strategy with uncertainty-based dynamic weighting is proposed to attach classification heads at all network stages. A temperature-regulated cross-stage mutual-distillation mechanism is further designed to enable bidirectional knowledge transfer among classification heads at different stages. On three benchmark HSI datasets, SGGCMNet achieves state-of-the-art accuracy. Full article

Background: Lateral epicondylopathy is commonly approached as a tendinopathic disorder of the common extensor origin; however, persistent lateral elbow pain may also involve a superficial sensory nerve component related to the posterior antebrachial cutaneous nerve (PABCN). This proof-of-concept cadaveric anatomical feasibility study evaluated whether a single-window, humerus-parallel ultrasound-guided injectate pathway could simultaneously reach the superficial PABCN-related fascial plane and the common extensor origin. Methods: One fresh-frozen male cadaveric donor was used, and both elbows were injected under real-time ultrasound guidance. With the elbow flexed and the forearm pronated, the transducer was aligned parallel to the long axis of the humerus over the lateral epicondylar region. A 23-gauge, 6 cm needle was advanced in plane from distal to proximal over the common extensor aponeurosis, and 10 mL of 1% methylene blue was injected into each elbow. Layer-by-layer anatomical dissection was then performed by an anatomist who was not involved in the injection procedure. Gross linear dye spread was measured directly during dissection using the distal needle entry point as the reference point, and ruler-containing photographs were additionally reviewed using ImageJ software for supportive image-assisted assessment. Results: In both elbows, methylene blue stained the superficial PABCN-related fascial plane, including the anterior and posterior branches of the PABCN, and concomitantly covered the common extensor aponeurosis and lateral epicondylar enthesis. Dye spread measured approximately 10 cm proximally, 5 cm distally, and 4 cm anteriorly. No gross intra-articular dye deposition or focal intramuscular pooling was observed. Conclusions: This proof-of-concept cadaveric study demonstrates the anatomical plausibility of a single-window, enthesis-centered ultrasound-guided injectate pathway that includes both the superficial PABCN-related plane and the common extensor origin. These findings should be interpreted as descriptive anatomical feasibility observations and do not establish reproducibility across anatomical variants, clinical efficacy, safety, or procedural superiority. Full article

In this paper a new mathematical optimization model is proposed to optimize the number of data aggregators in smart grids and assign each smart meter to at least one data aggregator. The model is based on a Set-Covering Problem, which aims to find the minimum number of sets that cover all elements. In the case of smart grids, these elements will be the smart meters. We used a branch–bound algorithm for the new optimization model to solve several instances considering different smart grid scenarios. In the scenarios, real-world parameters were used for a lot of smart meters (which ranged from 15 to 900, with output powers of 23 and 30 dBm used in the theoretical analysis), data aggregator costs, dispersions, maximum budget, and signal propagation losses. The tests reached the best values for the objective function with small, medium and large-scale instances in low computational times. Theoretical analysis was used to evaluate the signal received from data aggregators, indicating that they can receive information with a high-quality signal.The proposed model provides insights for stakeholders involved and can aid in smart grid implementation. In addition, it can offer a blueprint for engineers to optimize data flow within hierarchical grid structures with smart meters and data aggregators. Full article

Optical and Synthetic Aperture Radar (SAR) image registration is a fundamental task in remote sensing information fusion, yet it remains challenging due to significant differences in imaging mechanisms, radiation characteristics, and noise properties between the two modalities. Existing public datasets suffer from limited resolution, small scale, and insufficient scene diversity, and these limitations have hindered algorithm development. This paper constructs a large-scale, high-resolution optical–SAR registration dataset based on the HongTu-1 satellite 3-m SAR imagery and Google Earth optical imagery at zoom level 17, covering diverse scenes across China with a standardized pipeline including terrain correction, geometric alignment, standardized slicing, and quality filtering. Building upon this dataset, a hand-crafted keypoint-based cross-modal registration method is proposed, incorporating multi-level edge filtering and hybrid feature detection. Unlike conventional hand-crafted methods such as RIFT, SRIF, and LNIFT, which mainly refine keypoint detection, description, or matching within a SIFT-style pipeline, the core novelty of this work lies in SAR-specific preprocessing and multi-level hybrid filtering. These components are designed to suppress speckle while extracting more stable and discriminative shared edge responses for cross-modal registration. An improved Log-domain Total Variation (Log-TV) denoising model is introduced for SAR preprocessing. A hybrid edge filtering framework combining phase congruency analysis and Structured Random Forest (SRF) edge detection is constructed within a Gaussian scale space. A dual-branch feature detection scheme integrating blob and corner features is designed with a robust orientation assignment strategy. Feature description uses the Gradient Location–Orientation Histogram (GLOH) descriptor with Principal Component Analysis (PCA) reduction, while geometric estimation employs the Fast Sample Consensus (FSC) algorithm. Experiments on the self-constructed HT dataset and on the public OSdataset and SAR2Opt benchmarks show that the proposed method consistently achieves low RMSE and high success rates. It also maintains competitive efficiency among hand-crafted methods while retaining strong robustness to scale and rotation variations. Full article

Underground duct cables are widely used in urban power distribution systems, but their enclosed installation environment makes defect inspection difficult, labor-intensive, and potentially hazardous. Infrared thermography can capture abnormal temperature distributions caused by insulation degradation, conductor damage, sheath failure, or severe structural defects, while robot-based inspection provides a promising solution for confined duct environments. However, thermographic fault detection for underground small-diameter duct cables remains insufficiently studied, and practical deployment requires lightweight models suitable for embedded edge devices. In this study, an improved YOLO26-based thermographic fault detection framework is proposed for underground duct cable inspection. A Cable-Thermo dataset is constructed using an ANSYS 2025 R2-based thermoelectric coupling simulation, covering four defect categories: hollow-type damage, conductor burnout, sheath damage, and severe damage. To balance detection accuracy and deployment efficiency, two model variants are developed. YOLO26-Thermo-E retains the original detection scales and integrates CDA and SimSPPF modules for accuracy-prioritized diagnosis. YOLO26-Thermo-H further removes the small-scale detection branch as a deployment-oriented design choice, based on the scale distribution observed in the simulation dataset, where most fault-induced thermal anomalies appear as spatially continuous medium- or large-scale regions. This design assumption still requires further validation using real duct thermographic data. Experiments show that YOLO26-Thermo-E achieves the highest mAP50 of 99.20%. YOLO26-Thermo-H maintains a mAP50 of 99.00% while reducing GFLOPs by 34.3% and parameters by 16.2% compared with YOLO26. On an NVIDIA Jetson Orin NX, YOLO26-Thermo-H reaches 34 FPS under FP16 inference and 45 FPS under INT8 inference. These results demonstrate the feasibility of the proposed framework under controlled simulation conditions and its potential for edge deployment. The limitations of the simulation-based dataset are also discussed, and future work will focus on real-scene data collection and simulation-to-real generalization. Full article

Frequency-hopping (FH) signals appear as small rectangular pulses in time-frequency spectrograms. At low signal-to-noise ratios (SNRs), noise along the frequency axis, caused by short-time Fourier transform (STFT) spectral leakage, blurs pulse boundaries, while the varying scales of hop rectangles exceed the capacity of a single receptive field. This paper presents YOLO11-FH, a modified YOLO11 detector that introduces two signal-processing-motivated modules. A FreqSmoothBlock (FSB) uses a $(3,1)$ depthwise convolution to smooth exclusively along the frequency axis, while adding only $5C$ parameters. A TFMultiResBlock (TFMRB) fuses three parallel dilated convolution branches (dilation rates of 1, 2, and 3) to cover different hop scales, replacing a heavier C3k2 module. The detection head is further simplified by halving the Bottleneck repeat count and disabling the deep submodule at the P5 scale. On a simulated FH dataset (SNRs ranging from $-15$ dB to $-10$ dB, five jamming types), YOLO11-FH achieves 96.04% mean average precision (mAP)@0.5 and 76.18% mAP@0.5:0.95, outperforming the YOLO11n baseline by 0.95 and 2.91 percentage points (pp) with 2.9% fewer parameters. Full article

Hyperspectral image classification remains challenging because robust recognition requires preserving spatial–spectral coupling, extracting complementary spectral and spatial cues, and fusing heterogeneous features without excessive redundancy. To address this issue, a dual-branch convolutional neural network (CNN) with depthwise separable fusion, termed DSFA-CNN, is developed. The network combines a 3D convolution branch for coupled spatial–spectral representation learning with a 1D+2D branch for efficient spectral and spatial modeling. A convolutional block attention module (CBAM) is introduced in the decomposed branch to emphasize informative spectral responses and salient spatial regions, and a depthwise separable fusion module is used to improve cross-branch integration while limiting fusion-stage redundancy and the risk of overfitting. Experiments on Indian Pines, University of Pavia, Salinas, and Houston2013 yield overall accuracies of 95.62 ± 0.13%, 99.25 ± 0.13%, 99.89 ± 0.11%, and 97.62 ± 0.23%, respectively. The gains are most evident on the more challenging Indian Pines and Houston2013 scenes. Ablation results show that the dual-branch design provides complementary information, whereas CBAM and the fusion module further improve representation selectivity and feature integration. Computational cost analysis further indicates that DSFA-CNN achieves a more favorable trade-off between classification accuracy and computational efficiency than several recent competitive baselines. These results demonstrate the effectiveness of parallel coupled–decomposed modeling with efficient feature fusion for robust hyperspectral image classification. Full article