Search Results (3,796)

In complex maritime environments, the safety of path planning for Unmanned Surface Vehicles (USVs) remains a significant challenge. Existing methods for handling dynamic obstacles often suffer from inadequate predictability and generate non-smooth trajectories. To address these issues, this paper proposes a reliable hybrid path planning approach that integrates a reinforcement learning-enhanced A* algorithm with an improved Dynamic Window Approach (DWA). Specifically, the A* algorithm is augmented by incorporating a dynamic five-neighborhood search mechanism, a reinforcement learning-based adaptive weighting strategy, and a path post-optimization procedure. These enhancements collectively shorten the path length and significantly improve trajectory smoothness. While ensuring that the global path avoids dynamic obstacles smoothly, a Kalman Filter (KF) is integrated into the Long Short-Term Memory (LSTM) network to preprocess historical data. This mechanism suppresses transient outliers and stabilizes the trajectory prediction of dynamic obstacles. Moreover, the evaluation function of the DWA is refined by incorporating the International Regulations for Preventing Collisions at Sea (COLREGs) constraints, enabling compliant navigation behaviors. Simulation results in MATLAB demonstrate that the enhanced A* algorithm better conforms to the kinematic model of the USVs. The improved DWA significantly reduces collision risks, thereby ensuring safer navigation in dynamic marine environments. Full article

Accurate downhole depth measurement is essential for oil and gas well operations, directly influencing reservoir contact, production efficiency, and operational safety. Collar correlation using a casing collar locator (CCL) is fundamental for precise depth calibration. While neural network has achieved significant progress in collar recognition, preprocessing methods for such applications remain underdeveloped. Moreover, the limited availability of real well data poses substantial challenges for training neural network models that require extensive datasets. This paper presents a system integrated into a downhole toolstring for CCL log acquisition to facilitate dataset construction. Comprehensive preprocessing methods for data augmentation are proposed, and their effectiveness is evaluated using baseline neural network models. Through systematic experimentation across diverse configurations, the contribution of each augmentation method is analyzed. Results demonstrate that standardization, label distribution smoothing (LDS), and random cropping are fundamental prerequisites for model training, while label smoothing regularization (LSR), time scaling, and multiple sampling significantly enhance model generalization capabilities. Incorporating the proposed augmentation methods into the two baseline models results in maximum F1 score improvements of 0.027 and 0.024 for the TAN and MAN models, respectively. Furthermore, applying these techniques yields F1 score gains of up to 0.045 for the TAN model and 0.057 for the MAN model compared to prior studies. Performance evaluation on real CCL waveforms confirms the effectiveness and practical applicability of our approach. This work addresses the existing gaps in data augmentation methodologies for training casing collar recognition models under CCL data-limited conditions, and provides a technical foundation for the future automation of downhole operations. Full article

Background/Objectives: Parkinsonian gait is characterized by impaired motor adaptability, altered temporal organization, and reduced movement variability. While data augmentation is commonly used to mitigate class imbalance in gait-based machine learning models, conventional strategies often ignore physiological differences between healthy and pathological movements, potentially distorting meaningful motor dynamics. This study explores whether preserving healthy motor variability while selectively augmenting pathological gait signals can improve the robustness and physiological coherence of gait pattern classification models. Methods: Eight patients with Parkinsonian gait patterns and forty-eight healthy participants performed walking tasks on the Motigravity platform under hypogravity conditions. Full-body kinematic data were acquired using wearable inertial sensors. A selective augmentation strategy based on smooth time-warping was applied exclusively to pathological gait segments (×5, σ = 0.2), while healthy gait signals were left unaltered to preserve natural motor variability. Model performance was evaluated using a hybrid convolutional neural network–long short-term memory (CNN–LSTM) architecture across multiple augmentation configurations. Results: Selective augmentation of pathological gait signals achieved the highest classification performance (94.1% accuracy, AUC = 0.97), with balanced sensitivity (93.8%) and specificity (94.3%). Performance decreased when augmentation exceeded an optimal range of variability, suggesting that beneficial augmentation is constrained by physiologically plausible temporal dynamics. Conclusions: These findings demonstrate that physiology-informed, selective data augmentation can improve gait pattern classification under constrained data conditions. Rather than supporting disease-specific diagnosis, this proof-of-concept study highlights the importance of respecting intrinsic differences in motor variability when designing augmentation strategies for clinical gait analysis. Future studies incorporating disease-control cohorts and subject-independent validation are required to assess specificity and clinical generalizability. Full article

Prostate cancer is diagnosed through a comprehensive study of histopathology slides, which takes time and requires professional interpretation. To minimize this load, we developed a semi-supervised learning technique that combines transformer-based representation learning and a custom TransUNet classifier. To capture a wide range of morphological structures without manual annotation, our method pretrains DINOv2 on 10,000 unlabeled prostate tissue patches. After receiving the transformer-derived features, a bespoke CNN-based decoder uses residual upsampling and carefully constructed skip connections to merge data from many spatial scales. Expert pathologists identified only 20% of the patches in the whole dataset; the remaining unlabeled samples were contributed by using a consistency-driven learning method that promoted reliable predictions across various augmentations. The model received precision and recall scores of 91.81% and 89.02%, respectively, and an accuracy of 93.78% on an additional test set. These results exceed the performance of a conventional U-Net and a baseline encoder–decoder network. All things considered, the localized CNN (Convolutional Neural Network) decoding and global transformer attention provide a reliable method for prostate cancer classification in situations with little annotated data. Full article

The proliferation of Internet of Things (IoT) devices has created unprecedented security challenges characterized by resource constraints, heterogeneous network architectures, and severe class imbalance in attack detection datasets. This paper presents a comprehensive benchmark evaluation of five Generative Adversarial Network (GAN) architectures for energy-aware intrusion detection: Standard GAN, Progressive GAN (PGAN), Conditional GAN (cGAN), Graph-based GAN (GraphGAN), and Wasserstein GAN with Gradient Penalty (WGAN-GP). Our evaluation framework introduces novel energy-normalized performance metrics, including Accuracy-per-Joule (APJ) and F1-per-Joule (F1PJ), that enable principled architecture selection for energy-constrained deployments. We propose an optimized WGAN-GP architecture incorporating diversity loss, feature matching, and noise injection mechanisms specifically designed for classification-oriented data augmentation. Experimental results on a stratified subset of the BoT-IoT dataset (approximately 1.83 million records) demonstrate that our optimized WGAN-GP achieves state-of-the-art performance, with 99.99% classification accuracy, a 0.99 macro-F1 score, and superior generation quality (MSE 0.01). While traditional classifiers augmented with SMOTE (i.e., Logistic Regression and CNN1D-TCN) also achieve 99.99% accuracy, they suffer from poor minority class detection (77.78–80.00%); our WGAN-GP improves minority class detection to 100.00% on the reported test split (45 of 45 attack instances correctly identified). Furthermore, WGAN-GP provides substantial efficiency advantages under our energy-normalized metrics, achieving superior accuracy-per-joule performance compared to Standard GAN. Also, a cross-dataset validation across five benchmarks (BoT-IoT, CICIoT2023, ToN-IoT, UNSW-NB15, CIC-IDS2017) was implemented using 250 pooled test attacks to confirm generalizability, with WGAN-GP achieving 98.40% minority class accuracy (246/250 attacks detected) compared to 76.80% for Classical + SMOTE methods, a statistically significant 21.60 percentage point improvement ($p<0.0001$). Finally, our analysis reveals that incorporating diversity-promoting mechanisms in GAN training simultaneously achieves best generation quality AND best classification performance, demonstrating that these objectives are complementary rather than competing. Full article

Steel surface defect detection is essential for maintaining industrial production quality and operational safety. However, existing deep learning-based methods often encounter high computational costs, hindering their deployment on mobile devices. To effectively address this challenge, we propose a lightweight and efficient steel surface defect detection network based on feature segmentation and partially connected structures, termed LESSDD-Net. In LESSDD-Net, we first introduce a lightweight downsampling module called the cross-stage partial-based dual-branch downsampling module (CSPDDM). This module significantly reduces the number of model parameters and computational costs while facilitating more efficient downsampling operations. Next, we present a lightweight attention mechanism known as coupled channel attention (CCAttention), which enhances the model’s capability to capture essential information in feature maps. Finally, we improve the faster implementation of cross-stage partial bottleneck with two convolutions (C2f) and design a lightweight version called the lightweight and partial faster implementation of cross-stage partial bottleneck with two convolutions (LP-C2f). This module not only enhances detection accuracy but also further diminishes the model’s size. Experimental results on the data-augmented Northeastern University surface defect detection (NEU-DET) dataset indicate that the mean average precision (mAP) of LESSDD-Net improves by 3.19% compared to the baseline model YOLO11n. Additionally, in terms of model complexity, LESSDD-Net reduces the number of parameters and computational costs by 39.92% and 20.63%, respectively, compared to YOLO11n. When compared with other mainstream object detection models, LESSDD-Net achieves top detection accuracy with the highest mAP value and demonstrates significant advantages in model complexity, characterized by the lowest number of parameters and computational costs. Full article

This study introduces a class-balanced Convolutional Neural Network (CNN) framework specifically designed for the binary classification of breast tumors in digital mammography. The proposed method systematically addresses the pervasive issue of class imbalance in medical imaging datasets by implementing advanced dataset balancing strategies, which resulted in a significant reduction in false negatives that is critical in early breast cancer detection. The proposed architecture is designed for high-resolution mammograms and employs regularization techniques, such as dropout and L2 weight decay, which are intended to enhance generalization and reduce the risk of overfitting. Comprehensive data augmentation and normalization further enhance the model’s robustness and adaptability to real-world clinical variability. Evaluated on the MIAS dataset, our balanced CNN achieved an accuracy of 98.84%, exhibiting both sensitivity and overall reliability. This work demonstrates that a class-balanced CNN can deliver both high diagnostic accuracy and computational efficiency, indicating potential for future use in clinical screening workflows. The system’s ability to minimize diagnostic errors and support radiologists with reliable, data-driven predictions represents an exploratory step toward improving automated breast cancer detection. Full article

Coalbed methane, an abundant clean energy resource in China, is gaining significant attention. Electric submersible progressive cavity pumps, ideal for downhole extraction with high solids content, are vital in coalbed methane operations. Current fault diagnosis research for these pumps mainly relies on machine learning algorithms to identify fault features, but complex working conditions and imbalanced sample distributions challenge these models’ ability to perceive multi-scale and multi-dimensional features. To enhance the model’s perception of deep abnormal data in complex multi-case industrial datasets, this study proposes a deep learning model based on a multi-scale extraction and residual module convolutional neural network. Innovatively, a cross-attention module using global autocorrelation and local cross-correlation is introduced to constrain the multi-scale feature extraction process, making the model better suited to specific and differentiated data environments. Post feature extraction, the model employs Borderline-SMOTE to augment minority class samples and uses Tomek Links for noise removal. These enhancements improve the comprehensive perception of fault types with significant differences in period, amplitude, and dimension, as well as the learning capability for rare faults. Based on field-collected fault data and using enhanced and cleaned features for classifier training, tests on a real industrial dataset show the proposed model achieves an F1 Measure of 90.7%—an improvement of 13.38% over the unimproved model and 9.15–31.64% over other common fault diagnosis models. Experimental results confirm the method’s effectiveness in adapting to extremely imbalanced sample distributions and complex, variable field data characteristics. Full article

Soil organic matter (SOM) is a key indicator for evaluating soil fertility and ecological functions, and hyperspectral technology provides an effective means for its rapid and non-destructive estimation. However, in practical soil systems, the spectral response of SOM is often highly covariant with mineral composition, moisture conditions, and soil structural characteristics. Under small-sample conditions, hyperspectral SOM modeling results are usually highly sensitive to spectral preprocessing methods, sample perturbations, and model architecture and parameter configurations, leading to fluctuations in predictive performance across independent runs and thereby limiting model stability and practical applicability. To address these issues, this study proposes a multi-strategy collaborative deep learning modeling framework for small-sample conditions (SE-EDCNN-DA-LWGPSO). Under unified data partitioning and evaluation settings, the framework integrates spectral preprocessing, data augmentation based on sensor perturbation simulation, multi-scale dilated convolution feature extraction, an SE channel attention mechanism, and a linearly weighted generalized particle swarm optimization algorithm. Subtropical red soil samples from Guangxi were used as the study object. Samples were partitioned using the SPXY method, and multiple independent repeated experiments were conducted to evaluate the predictive performance and training consistency of the model under fixed validation conditions. The results indicate that the combination of Savitzky–Golay filtering and first-derivative transformation (SG–1DR) exhibits superior overall stability among various preprocessing schemes. In model structure comparison and ablation analysis, as dilated convolution, data augmentation, and channel attention mechanisms were progressively introduced, the fluctuations of prediction errors on the validation set gradually converged, and the performance dispersion among different independent runs was significantly reduced. Under ten independent repeated experiments, the final model achieved R² = 0.938 ± 0.010, RMSE = 2.256 ± 0.176 g·kg⁻¹, and RPD = 4.050 ± 0.305 on the validation set, demonstrating that the proposed framework has good modeling consistency and numerical stability under small-sample conditions. Full article

The lack of clinical data for chronic kidney disease (CKD) prediction frequently results in model overfitting and inadequate generalization to novel samples. This research mitigates this constraint by utilizing a Conditional Tabular Generative Adversarial Network (CTGAN) to enhance a constrained CKD dataset sourced from the University of California, Irvine (UCI) Machine Learning Repository. The CTGAN model was trained to produce realistic synthetic samples that preserve the statistical and feature distributions of the original dataset. Multiple machine learning models, such as AdaBoost, Random Forest, Gradient Boosting, and K-Nearest Neighbors (KNN), were assessed on both the original and enhanced datasets with incrementally increasing degrees of synthetic data dilution. AdaBoost attained 100% accuracy on the original dataset, signifying considerable overfitting; however, the model exhibited enhanced generalization and stability with the CTGAN-augmented data. The occurrence of 100% test accuracy in several models should not be interpreted as realistic clinical performance. Instead, it reflects the limited size, clean structure, and highly separable feature distributions of the UCI CKD dataset. Similar behavior has been reported in multiple previous studies using this dataset. Such perfect accuracy is a strong indication of overfitting and limited generalizability, rather than feature or label leakage. This observation directly motivates the need for controlled data augmentation to introduce variability and improve model robustness. The dataset with the greatest dilution, comprising 2000 synthetic cases, attained a test accuracy of 95.27% utilizing a stochastic gradient boosting approach. Ensemble learning techniques, particularly gradient boosting and random forest, regularly surpassed conventional models like KNN in terms of predicted accuracy and resilience. The results demonstrate that CTGAN-based data augmentation introduces critical variability, diminishes model bias, and serves as an effective regularization technique. This method provides a viable alternative for reducing overfitting and improving predictive modeling accuracy in data-deficient medical fields, such as chronic kidney disease diagnosis. Full article

This paper proposes a PI-Dual-STGCN fault diagnosis model based on a SHAP-LLM joint explanation framework to address issues such as the lack of transparency in the diagnostic process of deep learning models and the weak interpretability of diagnostic results. PI-Dual-STGCN enhances the interpretability of graph data by introducing physical constraints and constructs a dual-graph architecture based on physical topology graphs and signal similarity graphs. The experimental results show that the dual-graph complementary architecture enhances diagnostic accuracy to 99.22%. Second, a general-purpose SHAP-LLM explanation framework is designed: Explainable AI (XAI) technology is used to analyze the decision logic of the diagnostic model and generate visual explanations, establishing a hierarchical knowledge base that includes performance metrics, explanation reliability, and fault experience. Retrieval-Augmented Generation (RAG) technology is innovatively combined to integrate model performance and Shapley Additive Explanations (SHAP) reliability assessment through the main report prompt, while the sub-report prompt enables detailed fault analysis and repair decision generation. Finally, experiments demonstrate that this approach avoids the uncertainty of directly using large models for fault diagnosis: we delegate all fault diagnosis tasks and core explainability tasks to more mature deep learning algorithms and XAI technology and only leverage the powerful textual reasoning capabilities of large models to process pre-quantified, fact-based textual information (e.g., model performance metrics, SHAP explanation results). This method enhances diagnostic transparency through XAI-generated visual and quantitative explanations of model decision logic while reducing the risk of large model hallucinations by restricting large models to reasoning over grounded, fact-based textual content rather than direct fault diagnosis, providing verifiable intelligent decision support for industrial fault diagnosis. Full article

Within the traditional electronic neural network framework, Generative Adversarial Networks (GANs) have achieved extensive applications across multiple domains, including image synthesis, style transfer and data augmentation. Recently, several studies have explored the use of optical neural networks represented by the diffractive deep neural network (D²NN) for GANs. However, most of these focus on applications of the generative network, and there is currently no well-established D²NN architecture that simultaneously implements generative adversarial functionality. Here, we propose a novel implementation scheme for generative adversarial networks based on all-optical diffraction layers, demonstrating a complete all-optical adversarial architecture that simultaneously realizes both the generative network and the adversarial network (D²NN-GAN). We validated this method on the MNIST handwritten digit dataset, achieving Nash equilibrium convergence with the discriminator accuracy stabilizing around 50%. Concurrently, the average SSIM parameter of generated images reached 0.9573, indicating that the generated samples possess high quality and closely resemble real samples. Furthermore, we extended the framework to the KTH human action dataset, successfully reconstructing the “running” action with a discriminator accuracy of approximately 75%. The D²NN-GAN architecture introduces a fully optical generative adversarial model, providing a practical path for future optical modeling methods, such as image generation and video synthesis. Full article

Passive acoustic monitoring (PAM) using hydrophones enables collecting acoustic data to be collected in large and diverse quantities, necessitating the need for a reliable automated classification system. This paper presents a data-processing pipeline and a set of neural networks designed for a humpback-whale-detection system. A collection of audio segments is compiled using publicly available audio repositories and extensively curated via manual methods, undertaking thorough examination, editing and clipping to produce a dataset minimizing bias or categorization errors. An array of standard data-augmentation techniques are applied to the collected audio, diversifying and expanding the original dataset. Multiple neural networks are designed and trained using TensorFlow 2.20.0 and Keras 3.13.1 frameworks, resulting in a custom curated architecture layout based on research and iterative improvements. The pre-trained model MobileNetV2 is also included for further analysis. Model performance demonstrates a strong dependence on both feature representation and network architecture. Mel spectrogram inputs consistently outperformed MFCC (Mel-Frequency Cepstral Coefficients) features across all model types. The highest performance was achieved by the pretrained MobileNetV2 using mel spectrograms without augmentation, reaching a test accuracy of 99.01% with balanced precision and recall of 99% and a Matthews correlation coefficient of 0.98. The custom CNN with mel spectrograms also achieved strong performance, with 98.92% accuracy and a false negative rate of only 0.75%. In contrast, models trained with MFCC representations exhibited consistently lower robustness and higher false negative rates. These results highlight the comparative strengths of the evaluated feature representations and network architectures for humpback whale detection. Full article

In the Internet of Things (IoT), ultra-wideband (UWB) plays an essential role in localization and navigation. However, in indoor environments, UWB signals are often blocked by obstacles, leading to non-line-of-sight (NLOS) propagation. Thus, reliable line-of-sight (LOS)/NLOS identification is essential for reducing errors and enhancing the robustness of localization. This paper focuses on a single-anchor UWB configuration and proposes a temporal deep learning framework that jointly exploits two-way ranging (TWR) and angle-of-arrival (AOA) measurements for LOS/NLOS identification. At the core of the model is a temporal convolutional network (TCN) augmented with a self-attentive pooling mechanism, which enables the extraction of dynamic propagation patterns and temporal contextual information. Experimental evaluations on real-world measurement data show that the proposed method achieves an accuracy of 96.65% on the collected dataset and yields accuracies ranging from 88.72% to 93.56% across the three scenes, outperforming representative deep learning baselines. These results indicate that jointly exploiting geometric and temporal information in a single-anchor configuration is an effective approach for robust UWB indoor positioning. Full article

With the rapid expansion of pulsar survey data driven by advanced radio telescopes such as FAST, automated detection methods have become crucial for the efficient and accurate identification of single-pulse signals. A key challenge in this task is the extreme class imbalance between genuine pulsar pulses and radio frequency interference (RFI), which significantly hampers classifier performance—particularly in low signal-to-noise ratio (S/N) environments. To address this issue and improve detection accuracy, we propose Pulsar-WRecon, a Wasserstein GAN with Gradient Penalty (WGAN-GP)-based framework designed to generate realistic single-pulse profiles. The synthetic samples generated by Pulsar-WRecon are used to augment training data and alleviate class imbalance. Building upon the enhanced dataset, Convolutional Kolmogorov–Arnold Network (CKAN) is further introduced as a novel hybrid model that integrates convolutional layers with KAN-based functional decomposition to better capture complex patterns in pulse signals. On the three-channel pulsar images from the HTRU1 dataset, our method achieves a recall of 97.5% and a precision of 98.5%. On the DM time series image dataset, FAST-DATASET, it achieves a recall of 93.2% and a precision of 92.5%. These results validate that combining generative data augmentation with an improved model architecture can effectively enhance the precision of single-pulse detection in large-scale pulsar surveys, especially in challenging, real-world conditions. Full article