Search Results (559)

Accurate detection of field insect pests remains a significant challenge for precision agriculture due to the elongated and variable morphology of the target organisms, their frequent resemblance to complex background textures, and the long-tail distribution of species in natural datasets. While deep convolutional neural networks (CNNs) have advanced the field, they are often constrained by a limited effective receptive field and the entanglement of semantic and spatial features, which can lead to elevated false-positive rates and missed detections for low-contrast or rare targets. This paper introduces a novel detection framework that integrates state space modeling with multi-stream feature disentanglement to address these limitations. First, a visual state space module is employed as the backbone feature extractor, enabling the establishment of a global receptive field with linear computational complexity and thereby improving the perception of long-range morphological structures. Second, a Topological Feature Disentanglement Pyramid Network is proposed. This architecture explicitly separates feature representations into semantic and spatial streams and recombines them through graph convolutional interactions, which serves to suppress background interference and enhance localization precision. A meta-auxiliary detection head, active only during training, is introduced to amplify supervision signals for hard, low-contrast samples via adversarial gradient modulation. Furthermore, an implicit neural radiance field augmentation pipeline is used to generate physically consistent synthetic views of underrepresented pest classes, mitigating the negative effects of long-tail data distributions. Experimental evaluations on the public BAU-Insectv2 benchmark demonstrate that the proposed method achieves a mean average precision (mAP@0.5) of 81.8%, representing a 4.4-percentage-point improvement over a comparable baseline, while maintaining a compact parameter count of 2.33 M and an inference speed of 178.6 FPS. The framework exhibits particular efficacy in detecting elongated, minute, and rare pests, suggesting a promising technical approach for real-time, field-based pest surveillance in precision agriculture. Full article

As tunnel engineering continues to advance toward deeper, longer, and more complex projects, the risks encountered during the construction phase have evolved into a combination of various disaster types and the accumulation of multiple contributing factors. Traditional empirical and semi-empirical risk management methods are increasingly revealing shortcomings in terms of timeliness, accuracy, and the ability to process multi-source data. In recent years, driven by advancements in computing power and sensor technology, artificial intelligence algorithms (AI algorithms) such as machine learning and deep learning have been rapidly adopted in tunnel construction risk management. This paper retrieved relevant literature from the Web of Science database covering the period from 2010 to 2025. After rigorous screening, 96 highly relevant papers were selected for bibliometric analysis. This paper systematically reviews research progress from two perspectives: algorithmic models and engineering applications. The review indicates that, in terms of algorithmic models, traditional machine learning, convolutional neural network, recurrent neural network, generative adversarial network, Transformer, and graph neural network constitute a multi-level technical framework encompassing feature representation, risk perception, and intelligent decision-making. In terms of applications, AI algorithms have been widely integrated into typical scenarios such as geological hazard identification and prediction, surrounding rock stability and deformation prediction, rock burst assessment and early warning, lining defect detection and structural safety assessment, construction-induced ground settlement prediction, and tunnel gas and fire hazard prediction, significantly enhancing risk identification and early warning capabilities. However, several challenges remain, including the scarcity of high-quality datasets, the prevalence of noisy, incomplete, and heterogeneous monitoring data, insufficient coupling between model interpretability and engineering mechanisms, limited cross-project transferability, and the lack of integrated management systems for multi-hazard lifecycle control. Based on this, this paper proposes future research directions in areas such as data infrastructure development, integration of mechanism constraints, and multi-hazard collaborative modeling, aiming to provide guidance for the further development of intelligent risk management in tunnel construction. Full article

In characterizing unconventional reservoirs, conventional Digital Rock Physics (DRP) has long been constrained by three fundamental bottlenecks: the trade-off between imaging resolution and field of view, challenges in reconstructing multiscale pore topology, and the prohibitive computational cost of direct numerical simulation (DNS) at the pore scale. The deep integration of artificial intelligence and rock physics has given rise to a new paradigm—Intelligent Digital Rock Physics (IDRP). This paper provides a systematic review of the evolutionary trajectory of IDRP, with a focus on how machine learning is reshaping the end-to-end workflow from imaging and segmentation to reconstruction and simulation. First, we survey image super-resolution and 3D pore structure generation techniques based on convolutional neural networks (CNNs), generative adversarial networks (GANs), and diffusion models, elucidating their mechanisms for surpassing optical diffraction limits and incorporating macroscopic petrophysical constraints. Second, we outline algorithmic strategies for fusing multi-source heterogeneous data (e.g., Micro-CT and SEM) and representing dual-porosity or multi-continuum systems. Third, we critically examine the application of machine learning surrogates in single- and multiphase flow prediction, highlighting how physics-informed machine learning (PIML) and reinforcement learning (RL)—by embedding governing equations such as Navier–Stokes or Muskat–Leverett into loss functions—achieve both computational acceleration and physical consistency. We further identify key limitations of current IDRP approaches, including insufficient validation of generated topological realism, narrow generalization across lithologies, inadequate representation of dynamic wettability, and limited model interpretability. Finally, we propose a forward-looking roadmap centered on multimodal foundation models for rocks, coupled with neural operators and uncertainty quantification frameworks, emphasizing the critical pathways for translating IDRP into engineering digital twins for unconventional hydrocarbon development, coalbed methane production enhancement, Enhanced Geothermal Systems, and geological CO₂ storage. This review offers a comprehensive reference for researchers at the intersection of geophysics, rock mechanics, and artificial intelligence. Full article

The following narrative review discusses the use of deep learning and 3D modeling in facial reconstruction from skeletal remains, focusing on accuracy, algorithmic bias, and evidential reliability. Forensic facial reconstruction (FFR) is a multidisciplinary field combining anthropology, medicine, and visual sciences to approximate the facial appearance of unidentified individuals from skeletal remains. Traditional manual methods, based on anatomical knowledge and facial soft tissue thickness (FSTT) measurements, are limited by subjectivity, labor intensity, and inter-expert variability. This narrative review summarizes contemporary AI-assisted approaches, with emphasis on convolutional neural networks (CNNs), generative adversarial networks (GANs), variational autoencoders (VAEs), and diffusion models, which enable probabilistic prediction of facial morphology while accounting for demographic variables such as sex, age, and population ancestry. Key challenges affecting reconstruction accuracy—including dataset limitations, population-specific variability, and algorithmic bias—are discussed, alongside quantitative validation methods and concerns regarding model transparency. Legal and ethical considerations, such as privacy, biometric data protection, and the need for explainable AI (XAI) frameworks, are highlighted. Future perspectives include hybrid expert–AI workflows, the development of globally representative datasets, and the integration of multimodal data sources, including DNA phenotyping, 3D morphometrics, and biomechanical modeling. These advances aim to create standardized, interpretable, and biologically informed frameworks that enable AI to support expert judgment and enhance the reliability of forensic facial reconstructions. Full article

Deep learning networks facilitate automated metal material/structural stress identification when employing the electromechanical impedance/admittance (EMI/EMA) of piezoelectric ceramic (PZT) transducers, while insufficient data quantity and low quality usually restrict the performance of data-driven deep networks. To address this problem, this paper innovatively proposed an original data enhancement method using the EMA generative adversarial network (EMAGAN) to overcome measurement data inefficiency and deficiency for deep learning-based stress identification, which is difficult to accomplish using the traditional EMA technique. In this method, a novel data-normalized algorithm was tuned to collaboratively foster the EMAGAN-based dataset generation. Then, the synthetic datasets incorporated with original ones were fed into an adaptively established one-dimensional convolutional neural network (1DCNN) for accurate stress prediction. A validating experiment was performed on an aluminum beam specimen subjected to uniaxial tensile load until failure, which was continuously monitored via two surface-bonded PZT transducers. The efficacy of the generated EMA datasets was evaluated through comparison with the raw ones in terms of statistical errors and deep learning-based aluminum structural stress identification. The results demonstrated that the EMAGAN generated high-accuracy EMA data which exceeded 380 times that of the normal collection method, and the EMAGAN paired with 1DCNN provides a promising way for EMA data-driven metal structural stress identification with high efficiency, intelligence and accuracy. Full article

The intelligent identification of grouting defects in grouted sleeves in prefabricated structures is critical for maintaining structural integrity. However, current deep learning-based identification methods face limitations, including insufficient model adaptability and the difficulty of obtaining labeled data. Models trained on one domain struggle to generalize to others due to differences in data distributions, making these methods challenging to apply in real-world scenarios. To address this engineering challenge, this paper investigates the applicability of maximum mean discrepancy-based domain adaptation (MMD-based DA) and domain adversarial training (DAT) approaches for cross-domain grouting defect identification. Acceleration signals collected by accelerometers near the grouted sleeves are used as the model input. The model’s ability to generalize across domains is evaluated by training on labeled data from one working condition and testing its performance on other working conditions using only unlabeled data. And these methods are compared with traditional Convolutional Neural Networks (CNNs). Experiments were conducted on a two-layer prefabricated frame structure. The experimental results demonstrated the effectiveness of the MMD-based DA method in improving the accuracy and robustness of defect identification across different domains, with the use of unlabeled data. Full article

Driven by the energy transition and carbon-neutrality targets, global demand for critical minerals is increasing rapidly, while the discovery of new mineral deposits has become increasingly challenging because easily detectable outcropping deposits are being depleted, and exploration is shifting toward concealed ore systems. In this context, data-driven approaches based on machine learning (ML) and deep learning (DL) are increasingly complementing conventional geological, geochemical, geophysical, and remote-sensing methods. This review provides a structured synthesis of AI-based mineral exploration studies published over the past decade, focusing on four key aspects: theoretical foundations; applications to diverse exploration datasets, including remote sensing, geochemistry, geophysics, and drill-core imagery; advances in mineral prospectivity mapping (MPM); and emerging trends and challenges, such as limited labeled data, uncertainty quantification, geological consistency, explainability, physics-informed neural networks (PINNs), and the adaptation of foundation models to geoscience data. Convolutional neural networks, autoencoders, generative adversarial networks, Transformers, and graph neural networks show strong potential for improving pattern recognition, data integration, and workflow automation. Overall, AI-based exploration is expected to play an increasingly important role in detecting concealed mineral deposits and strengthening resilient critical-mineral supply chains. Full article

Motor imagery (MI)-based brain–computer interfaces (BCIs) have led to great interest as a result of their potential use in neurorehabilitation, assistive robotics, and human–computer interaction. However, decoding electroencephalographic (EEG) signals with high accuracy continues to be a difficult task due to the weak signal-to-noise ratio, differences among subjects, and the complicated temporal–spectral–spatial neural dynamics. Deep learning methods recently developed, such as convolutional neural networks, recurrent architectures, graph neural networks, and adversarial transfer learning, have enhanced MI decoding performance, yet many models are still concentrating on a single representation domain or they need costly adaptation phases in terms of computation. To tackle these shortcomings, we present NeuroCrossNet, a unified tri-modal deep learning model that is able to learn the temporal, spectral, and spatial EEG features jointly for robust and calibration-free MI decoding. The suggested network combines a Temporal HyperMixer Block for capturing long-range temporal dependencies, a wavelet transformer for learning localized time–frequency representation, and a Graph Attention Network for EEG topology-aware spatial reasoning. Additionally, a Dynamic Residual Attention Gate (DRAG) has been developed to adaptively merge heterogeneous feature streams, and a compact subject-aware normalization (SAN) method enhances cross-subject generalization without the use of labeled target-domain calibration data. Our proposed model was tested following the rigorous leave-one-subject-out (LOSO) approach on BCI Competition IV-2a and High-Gamma datasets. NeuroCrossNet reached a classification accuracy of 91.30%, surpassing several strong benchmark methods, including CNN-LSTM, EEGNet, DeepConvNet, spectral CNN, and graph-based EEG decoding frameworks. Furthermore, a large number of ablation studies reveal that the integration of temporally, spectrally, and spatially complementary representations considerably boosts robustness and inter-subject consistency. Full article

With the advancement of remote sensing technology, image acquisition has become more convenient and the amount of information captured has significantly increased, playing a vital role in numerous fields. However, cloud cover often results in missing image data, severely affecting data usability. In recent years, although deep learning methods have made progress in cloud removal tasks, the complexity of modeling multispectral band relationships and the scarcity of paired data remain major challenges. To address this, this paper proposes a two-stage codebook fusion vector-quantized generative adversarial network (TCF-VQ GAN) and a training framework. The first stage employs synthetic aperture radar (SAR), MODIS, and cloud-free data for unsupervised training; the second stage performs fusion fine-tuning using SAR and MODIS on paired cloudy/cloud-free data. The model incorporates a space-channel jointed gated convolution (SCGC) module to model irregular cloud cover and combines channel attention for band selection, while a dynamically weighted wavelet alignment loss function (DW²A) is designed to enhance multiscale feature representation. Experiments on the SEN12MS-CR and SMILE-CR datasets demonstrate that the proposed method outperforms existing methods across all metrics: on SEN12MS-CR, PSNR is 31.0397 and SAM is 4.7243; they are 33.5191 and 2.1663, respectively, on SMILE-CR. Furthermore, under fixed paired data conditions, simply adding auxiliary and cloud-free data further improves performance, validating the method’s effectiveness in data-scarce scenarios. Full article

Sepsis is a prominent cause of death in intensive care units, and delayed diagnosis greatly worsens fatal outcomes due to the complex, irregular, and uneven character of clinical time-series data. Hence we proposed an interpretable and uncertainty-aware deep learning architecture that solves data quality, temporal irregularity, and clinical explainability restrictions, which are frequently addressed separately by existing models. The suggested method combines Bidirectional Recurrent Imputation for Time Series (BRITS)-based imputation, hybrid Conditional Tabular Generative Adversarial Network-Synthetic Minority Over-sampling Technique (CTGAN-SMOTE) data augmentation, a Temporal Convolutional Networks (TCN)-Attention architecture, and continuous-time neural Ordinary Differential Equations (ODEs), along with SHapley Additive exPlanations (SHAP)-based feature attribution and uncertainty quantification. The experimental evaluation on a large ICU dataset reveals greater predictive accuracy, with an AUROC of 0.926 and accurate early warnings up to six hours before clinical onset, all while maintaining strong interpretability and calibration. The proposed framework demonstrates strong predictive performance and provides early warnings up to six hours before clinical onset, while maintaining interpretability and calibration. While the results are promising, further validation across multiple clinical settings is required to confirm its generalisability and real-world applicability. Full article

Automated detection of defects in welds are inevitable in the assurance of structural integrity, but this faces serious challenges due to the microscopic characteristics of the discontinuities, low visual contrast and infrequent occurrence of defect samples. Conventional deep learning methods, while accurate, often lack interpretability and exhibit low recall for rare defects. This paper proposes a novel hybrid system combining a Generative Adversarial Network (GAN), a Convolutional Neural Network (CNN), and Extreme Gradient Boosting (XGBoost 2.0.0) to enhance weld defect classification performance and transparency. Firstly, a Deep Convolutional GAN (DCGAN) creates synthetic images of the minority classes; thus, the problem of class imbalance is resolved. Then, a pretrained ResNet50V2 CNN is used to extract features of the deep layers from the original images as well as from the generated ones. After that, these features are fed into an XGBoost classifier, which uses tree-based learning to optimize classification results and make the process more understandable to the user. Furthermore, interpretation is also facilitated by Grad-CAM rendering of the CNN regions of interest and SHAP analysis to measure the involvement of the features in XGBoost. Experiments using the available LoHi-WELD datasets show that the overall accuracy is significantly improved, the per-class recall of the rare defects is also enhanced, and the robustness is also improved. The proposed hybrid method not only achieves better results but also generates visual/explainable output, which is very valuable when the system is implemented in industrial welding inspection systems. This paper serves as a liaison between the latest AI technology and the practical interpretability requirements of the mechanical and welding engineering fields. Full article

Data imbalance significantly hinders the performance of deep learning models in rolling bearing fault diagnosis. While Generative Adversarial Networks (GANs) are widely used for data augmentation, traditional architectures employing pixel-level loss functions often fail to capture complex time-frequency textures, resulting in blurred spectrograms and the loss of transient fault characteristics. To address this, we propose a data augmentation framework based on a Perceptually Optimized Deep Convolutional GAN (PerDCGAN). By integrating a perceptual loss function derived from a pre-trained VGG-16 network, the generator is constrained at the feature level rather than the pixel level, explicitly enforcing the preservation of structural details and high-frequency impact patterns. Extensive experiments on the Case Western Reserve University (CWRU) and Paderborn University (PU) datasets demonstrate that the proposed method effectively mitigates spectral blurring. Ablation studies confirm the synergistic effect of the joint loss function. Furthermore, under extreme 0 dB noise conditions, the classifier augmented by PerDCGAN maintains a robust diagnostic accuracy of 89.65% on the PU dataset, significantly outperforming standard DCGAN and demonstrating strong potential for complex industrial applications. Full article

The geometric signatures of macroscopic interfaces in the two-dimensional critical Ising model strictly adhere to Schramm–Loewner Evolution (SLE) theory. In this study, we propose a physics-driven generative approach using Super-Resolution Generative Adversarial Networks (SRGANs) to approximate the inverse coarse-graining operation to generate larger configurations. From the perspective of Geometric Deep Learning (GDL), we leverage the geometric priors of Convolutional Neural Networks (CNNs)—specifically their translational and rotational symmetries—to effectively encode the universal physical laws of the Ising Hamiltonian. This inductive bias allows the model to be trained on small scales yet be generalized to large-scale systems ($2048 \times 2048$) while preserving physical conservation. To accommodate spin discreteness, we employ an L1-based loss function to maintain domain wall sharpness. SLE analysis and long-range correlation functions confirm that the model reproduces critical dynamics and conformal invariance, successfully serving as a physics-preserving inverse coarse-graining transformation framework. Full article

Electrical imaging logging images play a crucial role in petroleum exploration; however, in practical applications, blank strips frequently appear due to instrument malfunctions or data transmission failures, severely compromising geological interpretation and hydrocarbon evaluation. Existing image inpainting methods have limited adaptability to blank strips at different depth scales and exhibit blurred high-resolution geological textures. To address these issues, this paper proposes a blank strip filling method that integrates Arbitrary Kernel Convolution (AKConv) with the Aggregated Contextual-Transformations Generative Adversarial Network (AOT-GAN). Specifically, the adaptive sampling mechanism of AKConv is incorporated into the generator network of AOT-GAN, enabling the model—to effectively capture long-range contextual information and adaptively handle blank strips of varying scales and shapes through multi-scale feature fusion. Experimental results on real oilfield datasets demonstrate that the proposed method achieves significant improvements in PSNR, SSIM, and MAE, exhibiting superior structural preservation and texture sharpness—especially in restoring deep and large-scale blank strips. Furthermore, visual comparisons confirm the method’s superior performance in recovering key geological features, such as bedding continuity and fracture structures, thus providing an effective approach for electrical imaging logging image restoration. Full article

Remote sensing images acquired by UAVs under nighttime or low-illumination conditions suffer from insufficient illumination, leading to degraded image quality, detail loss, and noise, which restrict their application in public security and disaster emergency scenarios. Although existing machine learning-based enhancement methods can recover part of the missing information, they often cause color distortion and texture inconsistency. This study proposes an improved low-illumination image enhancement method based on a Weakly Paired Image Dataset (WPID), combining the Hierarchical Deep Convolutional Generative Adversarial Network (HDCGAN) with a low-rank image fusion strategy to enhance the quality of low-illumination UAV remote sensing images. First, YCbCr color channel separation is applied to preserve color information from visible images. Then, a Low-Rank Representation Fusion Network (LRRNet) is employed to perform structure-aware fusion between thermal infrared (TIR) and visible images, thereby enabling effective preservation of structural details and realistic color appearance. Furthermore, a weakly paired training mechanism is incorporated into HDCGAN to enhance detail restoration and structural fidelity. To achieve objective evaluation, a structural consistency assessment framework is constructed based on semantic segmentation results from the Segment Anything Model (SAM). Experimental results demonstrate that the proposed method outperforms state-of-the-art approaches in both visual quality and application-oriented evaluation metrics. Full article