Search Results (37,283)

Under the combined pressures of dual carbon policy constraints, the integration of intelligent construction technologies, and intensifying market competition, the development of a scientific and robust performance evaluation system has become essential for building construction enterprises seeking to enhance their core competitiveness. Traditional evaluation methods, however, often suffer from incomplete indicator systems and limited capability in addressing high-dimensional and nonlinear problems, rendering them inadequate for the evolving demands of the industry. To address these challenges, this study proposes a performance evaluation model for building construction enterprises based on the least squares support vector machine (LSSVM), optimized by an improved Pied Kingfisher Optimizer (IPKO). Drawing on environment–behavior theory, the model incorporates three environmental and ten behavioral factors. To overcome the limitations of the original PKO algorithm—namely, insufficient exploration capability and weak local search—the exploration phase of PKO is integrated with that of the Marine Predators Algorithm. Empirical results demonstrate that: (1) the proposed IPKO outperforms Particle Swarm Optimization (PSO), Whale Optimization Algorithm (WOA), Sparrow Search Algorithm (SSA), Dung Beetle Optimizer (DBO), Ospery Optimization Algorithm (OOA), and the original PKO in most benchmark functions; (2) the ReliefF feature selection algorithm improves the model’s test set accuracy by approximately 2.18%; and (3) the IPKO-LSSVM model achieves 6.53%, 4.16%, and 6.74% higher prediction accuracy than Backpropagation Neural Networks (BPNN), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost), respectively. These findings highlight the model’s effectiveness in addressing small-sample, high-dimensional, and nonlinear problems, offering a scientifically sound and practical tool for performance evaluation in building construction enterprises. Full article

The clinical diagnosis of sleep apnea conventionally necessitates resource-intensive Polysomnography (PSG). We propose a weakly supervised framework to detect apnea using non-invasive Ballistocardiography (BCG), thereby addressing the critical scarcity of labeled BCG data. Instead of manual annotation, our pipeline transfers knowledge from a synchronized ECG signal, using it as a “teacher” to generate pseudo-labels for the BCG model. We formulated a User-Defined Function (UDF) that combines Heart Rate Variability and ECG-Derived Respiration to autonomously label the BCG windows. These pseudo-labels were subsequently employed to train a 1D Convolutional Neural Network. Testing on a public dataset, the CNN model achieved 71.8% accuracy against the pseudo-labels. When projected against the clinical ground truth, we estimate a true accuracy of 77.7%. These results validate that ECG-based supervision can effectively train low-cost home sensors without the bottleneck of manual medical annotation. Full article

To address the uncertainty in fatigue life prediction of welded joints under small-sample conditions, this study proposes a prediction model based on support vector regression (SVR) enhanced by an improved Grey Wolf Optimizer (GWO). First, a CDE-GWO algorithm is developed by optimizing the convergence factor and integrating differential evolution (DE) to enhance population search ability; its effectiveness is verified via benchmark functions. Subsequently, a CDEGWO-SVR model is constructed and validated against SVR, GWO-SVR, DE-SVR, and DEGWO-SVR using UCI datasets, demonstrating superior fitting accuracy and lower error. Finally, the model is applied to aluminum welded joint fatigue data. Comparative analysis with radial basis function (RBF) neural networks and least squares S-N curve fitting across five evaluation metrics indicates that the proposed model achieves better performance in MSE, MAPE, R², and CC, with competitive RSD. Experimental results confirm that the CDEGWO-SVR model possesses stable and higher prediction precision, offering an effective solution for fatigue life prediction involving small samples and multiple uncertainty factors. Full article

Until now, there have been no published attempts to utilize ensemble learning approaches to pre-crash velocity estimation. In this research article, we focus on the method of vehicle crash velocity prediction based on the random forest regression approach. In particular, the study aims to develop and validate a random forest-based non-linear model for estimating pre-crash velocity using EES-related parameters for compact vehicles in a crash scenario against an immovable, stationary barrier. The estimation technique is trained and evaluated using the compact vehicle class from the NHTSA database, which consists of 399 records of frontal impacts against a rigid barrier. The relative error obtained for the presented calculation method is $7.57\mathrm{\%}$, with absolute error being equal to $1.12 \mathrm{m}/\mathrm{s}$. We subsequently compare our results with some other techniques which were tested on this dataset. Despite the simplicity of random forest regression, we obtain surprisingly good results, as the method outperforms linear regressor and artificial neural network predictors, which have relative errors of $8.17\mathrm{\%}$ and $9.63\mathrm{\%}$, respectively. The independence of Event Data Recorders along with the ease of obtaining the necessary data makes the proposed approach a highly desirable tool in forensic analysis, especially in cases involving older vehicles. Full article

Free space detection in assisted driving applications is essential to provide information to vehicles about traversable surfaces and potential obstacles to be avoided. The current trend in free space detection favors the use of deep learning techniques. However, Deep Neural Networks require extensive training that considers as many scenarios as possible, which makes it difficult to create a model that can be generalized to all types of surfaces. Additionally, their lack of explainability contrasts with the growing interest in geometrically grounded and safety-oriented design principles for autonomous vehicle systems. To address these limitations, we propose a geometric approach that incorporates coplanarity conditions and normal vector estimation, removing the dependence on datasets for different types of surfaces. Additionally, the stereoscopic images are clustered in superpixels. The use of images clustered in superpixels allows us to obtain shorter processing times, in addition to taking advantage of the spatial and color information provided by the superpixels to increase the robustness of the three-dimensional reconstruction of the scene. Experimental results show that the proposed superpixel-based approach achieves competitive performance compared to unsegmented dense stereo methods, while significantly reducing algorithmic complexity. These results demonstrate the viability of integrating superpixel clustering into stereo-based free space estimation frameworks. Full article

Wind-assisted ship propulsion (WASP) systems are gaining importance due to their contribution to reducing greenhouse gases and saving fuel. Existing studies mostly focus on the aerodynamics of sailing systems, the integration of sails and ship dynamics, and the prediction of fuel savings. The present study extends the use case of sailing systems by proposing a new control logic that improves maneuvering performance. Determining the spin ratio of rotor sails not only with thrust but also with side forces and moments is also included as an objective function. Using numerous random weights for each term and environmental conditions, the turning performance of the target ship was evaluated. Then, an artificial neural network (ANN) model was trained to decide on the optimal weights, depending on the environmental conditions. Finally, the performance of the new control approach was evaluated based on turning and zigzag test simulations. It was found that the advance, transfer, and tactical diameters dropped by up to 5%, 7% and 7%, respectively, compared to those of a conventional ship. When it comes to the zigzag performance, it was revealed that the overshoot angles dropped even though there was no simulation data about zigzag tests in the trained ANN model. Thus, it was shown that sails improve the maneuverability of ships in addition to providing additional thrust if a proper control approach is adopted. Full article

Event-based face detection has attracted significant interest due to the unique advantages of event cameras, including high temporal resolution, high dynamic range, and low power consumption. However, the lack of annotated public datasets remains a major challenge for training effective event-based face detection models. In this paper, we propose a spiking feature-driven synthetic event generation framework that utilizes a spiking neural network (SNN) in conjunction with a pretrained convolutional backbone to generate synthetic event representations from a single RGB image. To incorporate motion-induced ON/OFF polarity information, we introduce a movement-aware polarity integration (MPI) module that assumes four directional facial movements. An event-similarity score is further employed to select representations most consistent with real event data for training. Unlike conventional approaches relying on video-based simulators, our method enables efficient synthetic event dataset construction without requiring video inputs or additional simulation training. Experimental results on the N-Caltech101 dataset demonstrate a face detection accuracy of 99.91%, outperforming existing event-based face detection methods. Full article

Forecasting financial time series (FTS) is essential for analyzing and understanding the dynamics of financial markets. Traditional recurrent neural network (RNN) models often suffer from low prediction accuracy on non-stationary and abruptly changing data, as their gating mechanisms struggle to capture evolving trends in FTS. This paper introduces variational mode decomposition (VMD) and multifractal analysis to enhance the gating mechanism of the gated recurrent unit (GRU). By quantifying the changing characteristics of FTS, the proposed model dynamically adjusts the gating weights. In addition, a state fusion strategy is employed to improve the utilization efficiency of historical information. Experiments are conducted using daily data of the SSE 50, CSI 300, and CSI 1000 indices, spanning from 4 January 2002, to 26 December 2025. The results demonstrate that, compared to traditional models, the proposed model better captures the evolving characteristics of FTS and achieves higher prediction accuracy. Full article

Hydropower remains one of the primary sources of power generation. It can be operated as either a base-load or peak-load plant due to its rapid, easy start-up and stop-down capability. However, power plants, old or new, need to be operated and maintained optimally to meet energy demand and maximise economic returns. While the older plants without digital controls such as the Supervisory Control and Data Acquisition (SCADA) system are unable to leverage the evolving technology including big data and Artificial Intelligence (AI), the newer plants or plants that already have some form of data acquisition system have the advantage of leveraging the newer platforms for efficient operation, monitoring and fault diagnosis. Thus, an Artificial Neural Network (ANN), a machine learning (ML) algorithm, was chosen for this case study to predict the generator’s operational stator temperature by selecting six parameters that could potentially affect it. Real data from the 336 MW Chhukha Hydropower Plant (CHP) in Bhutan were used to train the ANN. The prediction of temperature using an ANN in MATLAB^® yielded an R² (correlation coefficient) of 96.8%, which is impressive but can be further improved through various optimisation and tuning methods with increased data volume and complexity. The performance of ANN prediction was validated against other regression models, and the ANN was found to outperform them. This demonstrated its capability to predict and detect generator temperature faults before failures, thereby enhancing hydropower operation and maintenance (O&M) efficiency. The model’s interpretation was also done through Shapley Additive ExPlanations (SHAP). Full article

Remaining Useful Life (RUL) prediction is a key component of prognostics and health management for modern industrial systems. While deep learning methods have significantly improved prediction accuracy, many existing approaches rely on large neural networks that are difficult to deploy on resource-constrained edge devices. This study presents a deployment-oriented evaluation of compact neural networks for RUL prediction using the NASA C-MAPSS turbofan engine benchmark. Two lightweight hybrid architectures, CNN–GRU and CNN–TCN, were developed with approximately 28k–32k parameters to represent realistic models for CPU-based edge inference. A systematic experimental analysis was conducted across all four C-MAPSS subsets (FD001–FD004), which represent increasing levels of operational and fault complexity. In addition to baseline performance, two post-training compression techniques (i.e., global unstructured magnitude pruning and dynamic INT8 quantization) were evaluated. To assess real deployment behavior, inference latency was measured on both a high-performance Intel x86 workstation and a resource-constrained ARM platform. Results show that CNN–GRU generally achieves higher predictive accuracy, whereas CNN–TCN provides more consistent and lower inference latency due to its convolution-only temporal modeling. Unstructured pruning can yield modest improvements in prediction accuracy, suggesting a regularization effect, but it does not reliably reduce model size or latency on standard CPUs due to the overhead associated with pruning masks. Dynamic quantization substantially reduces model size (particularly for CNN–GRU) while preserving predictive accuracy; however, it increases runtime latency because of additional quantization and dequantization operations. These findings demonstrate that compression techniques commonly used for large models do not necessarily translate into deployment benefits for already compact RUL architectures and highlight the importance of hardware-aware evaluation when designing edge prognostics systems. Full article

Age estimation based on facial images is a challenging task due to the complex and nonlinear nature of facial aging, which is influenced by both genetic and environmental factors. To address this challenge, we propose a hybrid ConvNeXt–Transformer framework that combines convolutional local feature extraction with attention-based global contextual modeling within a unified age regression pipeline. The methodological contribution of this work lies in the sequential integration of these two complementary paradigms for facial age estimation, allowing the model to capture both fine-grained textural cues—such as wrinkles and skin spots—and long-range spatial dependencies. We evaluate the proposed framework on benchmark datasets including MORPH II, CACD, UTKFace, and AFAD. The results show competitive performance across these datasets and confirm the effectiveness of the proposed hybrid design through extensive ablation analyses. Experimental results demonstrate that our approach achieves state-of-the-art MAE on MORPH II (2.26), CACD (4.35), and AFAD (3.09) under the adopted benchmark settings while remaining competitive on UTKFace. To address computational efficiency, we employ ImageNet pre-trained backbones and explore different architectural configurations, including fusion strategies and varying depths of the Transformer module, as well as regularization techniques such as stochastic depth and label smoothing. Ablation studies confirm the contribution of each component, particularly the role of attention mechanisms, in enhancing the model’s sensitivity to age-relevant features. Overall, the proposed hybrid framework provides a robust and accurate solution for facial age estimation, effectively balancing performance and computational cost. Full article

This study contributes to the modeling and optimization model for reverse osmosis water treatment (ROWT) due to a lack of economic and operational aspects. This study proposes a hybrid modeling and optimization framework using a hybrid artificial neural network (ANN) and genetic algorithm (GA) to enhance the accuracy of economic and operational predictions for ROWT. The ANN model is developed using seventeen key process parameters extracted from various ROWT plants, including flow rate, pH, conductivity, and turbidity. The GA is employed to optimize the network architecture and learning parameters based on the mean absolute percentage error (MAPE) as the fitness function. The findings of this study indicate that the GA-optimized model significantly outperforms the baseline model, reducing MAPE for the economic aspect (84.9% improvement) and the operational aspect (32.2% improvement). The findings from this study indicate that the hybrid ANN–GA approach is a management decision-making tool for reducing expenses without compromising water quality in ROWT management. The practical implications of this study are that predictions not only meet operational parameters but also predict expenses incurred, allowing managers to plan future budgets by optimizing ROWT resources and maintenance activities. Full article

Adversarial examples, created by adding small distortions to audio, can fool neural network models and cause automatic speech recognition (ASR) systems to produce incorrect outputs. Most current detection methods rely on the recognition capabilities of ASR systems. As a result, they often fail to detect such examples when ASR performance degrades or when facing an evasion attack—referred to in this paper as a “partial adversarial attack”—that is specifically designed to bypass ASR models. In this paper, we identify a distinct noise energy difference between adversarial examples and their original audio. Moreover, this noise energy difference is typically greater than that between adversarial examples and their re-attacked examples. This finding leads us to propose a novel detection method that fundamentally departs from traditional approaches by operating independently of ASR systems. The proposed method employs a detection strategy that involves re-attacking the input audio and identifies adversarial examples by characterizing the noise energy difference before and after re-attack without relying on ASR systems. The experimental results demonstrate that the proposed method detects various state-of-the-art adversarial attacks. Compared with the baselines, the proposed method achieves substantially better detection performance across standard adversarial examples, noisy adversarial examples, and partial adversarial examples. Full article

Non-destructive and automated detection of bacterial contamination is a critical prerequisite for ensuring high efficiency production and quality control in plant tissue culture. In this study, we developed a multispectral image acquisition system for Alocasia explants and proposed a novel image fusion model, termed Intensity-Texture enhanced Swin Fusion (ITSF). The ITSF framework employs convolutional neural networks to extract texture and intensity features from visible and near-infrared channels. Subsequently, a Swin Transformer-based module is integrated to model long-range spatial dependencies, ensuring cross-domain integration between the texture and intensity features. We formulated a composite loss function to guide the fusion process toward optimal results. This objective function integrates texture loss, entropy weighted structural similarity index (SSIM) and intensity aware dynamic gain guided loss. Experimental results demonstrate that the proposed method significantly enhances the visual saliency of bacteria and achieves superior quantitative performance across a comprehensive range of objective image fusion metrics. The detection performance reached a mean Average Precision (mAP50) of 0.949 with the fused images, satisfying industrial requirements for high-precision inspection, which provides a critical technical solution for the industrialization of automated micropropagation. Full article

The primary goal of manufacturing technologies in the metalworking industry is to provide products with specified quality characteristics, while maximizing time and cost efficiency. The total manufacturing cost in turning depends on a number of factors. The analysis of their effects and the estimation of the total manufacturing cost are of practical importance in process planning. Therefore, in the present study, the relationship between four inputs (depth of cut, feed rate, cutting speed and volume of material to be removed) and the total manufacturing cost in medium turning of C45E steel was modeled by using an artificial neural network (ANN). The developed ANN model was used for the analysis of the main and interaction effects of the aforementioned inputs on the total manufacturing cost. Verification of the observed effects was also carried out by applying the connection weight approach. The total manufacturing cost was mostly affected by depth of cut, while the effect of cutting speed was least pronounced. In addition, the results also revealed the presence of two-way interactions associated with cutting speed. For the given case study (with defined volume of material to be removed and specified machine tool), an optimized cutting regime was determined by developing and solving a single-objective turning optimization problem with three constraints related to chip slenderness, cutting power and depth of cut. Cutting force, needed for the estimation of cutting power, was estimated by using the dimensional analysis-based prediction model. Full article