Explore the vast range of titles available.

The strong heterogeneity of clastic reservoirs and the phenomenon of similar log responses for different lithologies (i.e., “same spectrum, different rocks”) significantly weaken feature separability. Furthermore, distribution shifts between different wells cause traditional models to suffer from severe generalization bottlenecks in cross-well applications. To address this critical challenge, this paper proposes a dual-driven framework comprising “Multivariate Feature Enhancement + Dynamic Ensemble”. At the feature level, physics-informed enhancement and multi-scale statistics are introduced to construct a Multivariate high-dimensional feature system, thereby strengthening the representation of geological patterns. At the model level, a sample-aware Dynamic Confidence-Weighted Ensemble (DCWE) strategy is designed to achieve sample-wise adaptive decision-making based on prediction uncertainty, fundamentally breaking through the limitations of fixed weights in static ensembles. This method combines the complementary advantages of Gradient Boosting Decision Trees (GBDT) and deep sequence networks, enabling the simultaneous capture of local textural variations and continuous trends across depths. Based on rigorous Leave-One-Group-Out (LOGO) cross-validation, the proposed framework achieves a maximum accuracy of 84.58%. It significantly reduces the misclassification rate in lithology transition zones and for minority class samples, while maintaining the geological continuity of prediction results. These results verify the significant advantages of the proposed method in cross-well generalization scenarios.

Total organic carbon (TOC) is a critical parameter for evaluating shale source rock quality and hydrocarbon generation potential. However, accurate TOC estimation from conventional well logs remains challenging, especially in data-limited geological settings. This study proposes an optimized XGBoost model for TOC prediction using conventional logging data from the Shahejie Formation in the Dongying Depression, Bohai Bay Basin, China. We systematically transform four standard logs—resistivity, acoustic transit time, density, and neutron porosity—into 165 candidate features through multi-scale smoothing, statistical derivation, interaction term creation, and spectral transformation. A two-stage feature selection process, combining univariate filtering and recursive feature elimination and further refined by principal component analysis, identifies ten optimal predictors. The model hyperparameters are optimized via Bayesian search within the Optuna framework to minimize cross-validation error. The optimized model achieves an R2 of 0.9395, with a Mean Absolute Error (MAE) of 0.3392, a Root Mean Squared Error (RMSE) of 0.4259, and a Normalized Root Mean Squared Error (NRMSE) of 0.0604 on the test set, demonstrating excellent predictive accuracy and generalization capability. This study provides a reliable and interpretable methodology for TOC characterization, offering a valuable reference for source rock evaluation in analogous shale formations and sedimentary basins.

In practical well-logging datasets, severe missing values, anomalous disturbances, and highly imbalanced lithology classes are pervasive. To address these challenges, this study proposes a well-logging lithology identification framework that combines Robust Feature Engineering (RFE) with quality-aware XGBoost. Instead of relying on interpolation-based data cleaning, RFE uses sentinel values and a meta-information tensor to explicitly encode patterns of missingness and anomalies, and incorporates sliding-window context to transform data defects into discriminative auxiliary features. In parallel, a quality-aware sample-weighting strategy is introduced that jointly accounts for formation boundary locations and label confidence, thereby mitigating training bias induced by long-tailed class distributions. Experiments on the FORCE 2020 lithology prediction dataset demonstrate that, relative to baseline models, the proposed method improves the weighted F1 score from 0.66 to 0.73, while Boundary F1 and the geological penalty score are also consistently enhanced. These results indicate that, compared with traditional workflows that rely solely on data cleaning, explicit modeling of data incompleteness provides more pronounced advantages in terms of robustness and engineering applicability.

Detecting surface defects in steel products is essential for maintaining manufacturing quality. However, existing methods struggle with significant challenges, including substantial defect size variations, diverse defect types, and complex backgrounds, leading to suboptimal detection accuracy. This work introduces ELS-YOLO, an advanced YOLOv11n-based algorithm designed to tackle these limitations. A C3k2_THK module is first introduced that combines a partial convolution, heterogeneous kernel selection protocoland the SCSA attention mechanism to improve feature extraction while reducing computational overhead. Additionally, the Staged-Slim-Neck module is developed that employs dual and dilated convolutions at different stages while integrating GMLCA attention to enhance feature representation and reduce computational complexity. Furthermore, an MSDetect detection head is designed to boost multi-scale detection performance. Experimental validation shows that ELS-YOLO outperforms YOLOv11n in detection accuracy while achieving 8.5% and 11.1% reductions in the number of parameters and computational cost, respectively, demonstrating strong potential for real-world industrial applications.

In recent years, due to the high number of unsafe factors in the construction industry, its death rate and injury rate have become the focus of research in the field of target detection. In the complex environment of the construction site, workers often forget to wear helmets or fail to wear them properly, causing hidden dangers for the safety of workers. Therefore, this paper proposes an application-oriented improvement strategy for helmet detection based on the YOLOv8 architecture (EC-YOLOv8), which integrates the ECA attention mechanism and content-sensing recombination feature operator into the YOLOv8 network, so as to further improve detection accuracy and detection time, and better meet the actual application requirements of construction sites. Enhanced Intersection over Union Loss (EIoU Loss) is introduced to improve the network model’s ability to evaluate the difference between the predicted boundary box and the real boundary box. The experimental results show that the EC-YOLOv8 algorithm proposed in this paper achieves 95.7% accuracy on the data set SHWD, which improves the detection accuracy of the helmet target.

Calcrete-type uranium prospectivity prediction is challenged by the strong heterogeneity of multi-source geoscientific raster datasets, weak anomaly responses, and the lack of explicit heterogeneous information organization in conventional deep learning models. In this study, the Yilgarn Craton of Western Australia was selected as the study area, and a geology-guided fixed-group fusion ResUNet model (GGF-ResUNet) was developed based on 12-channel multi-source geoscientific raster datasets. At the input stage, the evidence layers were divided into four fixed geoscientific proxy groups according to their data modality and geological interpretation, namely gravity, aeromagnetic, radiometric, and geochemical groups, and intra-group channel weighting together with inter-group gating was introduced to enhance the hierarchical representation and adaptive fusion of heterogeneous information. Ablation results showed that GGF-ResUNet achieved better performance than the baseline ResUNet, with AUC increasing from 0.9340 to 0.9740 and F1-score improving from 0.7264 to 0.8356. Further comparative experiments with Attention U-Net, U-Net, SegNet, and FCN showed that GGF-ResUNet achieved comparatively better quantitative performance and more spatially coherent prediction results under the current experimental setting. Without substantially increasing model complexity, the proposed method improves the representation and integration of heterogeneous geoscientific information and provides a feasible technical pathway for calcrete-type uranium prospectivity prediction under weak-anomaly conditions.

Knee osteoarthritis (KOA) is a leading cause of limited mobility and physical disability among the elderly. Early detection and intervention are crucial for slowing disease progression and improving patients’ quality of life. This paper proposes an automated diagnostic algorithm based on an improved YOLOv8s model, named KOA-YOLOv8s, to enhance the detection performance for KOA. The algorithm introduces an Efficient Convolutional Attention Module (ECAM), which employs the Efficient Channel Attention (ECA) mechanism to enhance the channel attention capabilities of the Convolutional Block Attention Module (CBAM). This enhancement enables the network to focus more effectively on the critical information within images, thereby improving detection accuracy. We designed an Improved Large Selective Kernel Focal Modulation module (LSK-FM) based on the large selective kernel network (LSK) and Focal Modulation module to replace the traditional Simplified Spatial Pyramid Pooling-Fast (SPPF) module. LSK enables the Focal modulation module to better focus on the details of the target. Additionally, we propose a lightweight detection head, named tiny-Head, which is based on depthwise separable convolution (DSC). This head replaces the two standard convolutional layers in the original detection head, reducing the parameter count and computational complexity. KOA-YOLOv8s achieved a mean Average Precision (mAP) of 79.69 % on a dataset consisting of 9786 X-ray images for KOA, which represents a 3.11 % improvement over the previous iteration, with a detection speed of 78.81 frames per second (FPS). The experimental results demonstrate that the detection performance of the proposed algorithm is comparable to that of other state-of-the-art algorithms.

In HEVC, the structure of coding unit (CU) and prediction unit (PU) is defined, which brings about higher coding efficiency than H.264/AVC. However, the rate distortion (RD) cost calculations of all depths of CUs and partition modes have yielded tremendous coding computational complexity. In order to reduce the complexity, a fast inter-prediction algorithm is proposed based on temporal and spatial correlations in this paper. In the proposed algorithm, the optimal partition mode in HEVC is selected based on its occurrence probability among all the partition modes and the similarity of the CU segmentation and partition mode between two adjacent frames is counted. Based on the optimal mode and similarity of CU segmentation and partition mode, at most two partition modes are evaluated for each CU to save computational complexity. In addition, the spatial correlation of CU segmentation and partition mode between the corresponding located (co-located) CU and its four surrounding CUs are analyzed. Based on this spatial correlation, only the optimal partition mode is evaluated to save the computational complexity for the deeper depths of CUs of the current CU. Simulation results show that the proposed algorithm achieves 65 % coding time reduction with negligible loss in coding efficiency and peak signal-to-noise ratio (PSNR), compared to previous fast mode decision algorithm.

A fast inter-prediction algorithm based on matching block features is proposed in this article. The position of the matching block of the current CU in the previous frame is found by the motion vector estimated by the corresponding located CU in the previous frame. Then, a weighted motion vector computation method is presented to compute the motion vector of the matching block of the current CU according to the motions of the PUs the matching block covers. A binary decision tree is built to decide the CU depths and PU mode for the current CU. Four training features are drawn from the characteristics of the CUs and PUs the matching block covers. Simulation results show that the proposed algorithm achieves average 1.1% BD-rate saving, 14.5% coding time saving and 0.01-0.03 dB improvement in peak signal-to-noise ratio (PSNR), compared to the present fast inter-prediction algorithm in HEVC.

Low power electronics endowed with artificial intelligence and biological afferent characters are beneficial to neuromorphic sensory network. Highly distributed synaptic sensory neurons are more readily driven by portable, distributed, and ubiquitous power sources. Here, we report a contact-electrification-activated artificial afferent at femtojoule energy. Upon the contact-electrification effect, the induced triboelectric signals activate the ion-gel-gated MoS 2 postsynaptic transistor, endowing the artificial afferent with the adaptive capacity to carry out spatiotemporal recognition/sensation on external stimuli (e.g., displacements, pressures and touch patterns). The decay time of the synaptic device is in the range of sensory memory stage. The energy dissipation of the artificial afferents is significantly reduced to 11.9 fJ per spike. Furthermore, the artificial afferents are demonstrated to be capable of recognizing the spatiotemporal information of touch patterns. This work is of great significance for the construction of next-generation neuromorphic sensory network, self-powered biomimetic electronics and intelligent interactive equipment. Low power electronics endowed with artificial intelligence and biological afferent characters are beneficial to neuromorphic sensory network. Here, the authors report contact-electrification-activated artificial afferent at femtojoule energy, which is able to carry out spatiotemporal recognition on external stimuli.