Search Results (32,484)

Defect detection in the hot rolling process is closely linked to the quality of the final product. Among these defects, slab camber during the intermediate rolling stage is one of the primary manifestations of asymmetry, which significantly impairs both the quality of the finished strip and the stability of subsequent rolling processes. Conventional image-based edge detection methods for slab camber are prone to detection deviations in complex industrial environments, mainly due to their weak noise robustness. To address the scientific challenge of low accuracy and poor robustness in feature extraction for hot-rolled intermediate slab camber detection, which is induced by environmental interference in complex industrial settings, we break through the technical bottlenecks of traditional edge detection methods and existing deep learning models in terms of channel–spatial feature collaborative optimization and anti-interference fusion of multi-scale features. We establish a dense perception network model integrated with a channel–spatial attention mechanism, realize robust feature recognition of slab edges under complex working conditions, and provide theoretical and technical support for the real-time quantitative detection of slab shape defects in the hot rolling process. The proposed model significantly improves detection accuracy and robustness through multi-scale feature enhancement and noise suppression, effectively meeting the requirements for real-time quantitative detection of slab camber in the roughing rolling stage. Field experiments verify that the method increases detection accuracy by 36.55% and achieves favorable performance on evaluation metrics, including ODS and OIS. Full article

Financial time series forecasting remains a challenging task due to the high non-stationarity and concept drift inherent to market data. Existing deep learning models, such as LSTMs and transformers, typically employ static weights after training, limiting their ability to adapt to rapid market regime shifts (e.g., from trends to reversals). To bridge this gap between static parameters and dynamic environments, we propose a novel framework named Game-Theoretic HyperNetwork (GTH-Net), which introduces a context-aware meta-learning mechanism to achieve adaptive forecasting. Specifically, we first introduce an Evolutionary Game-Theoretic Correction Module (E-GTCM) to explicitly extract latent buying and selling pressure based on market microstructure priors through an iterative gated evolution process. Subsequently, we propose a HyperNetwork-based fusion mechanism that treats the extracted game state as a meta-context to dynamically generate the weights of the forecasting head. This allows the model to automatically switch its prediction rules in response to shifting market regimes. Extensive experiments on real-world stock datasets demonstrate that GTH-Net significantly outperforms baselines in terms of machine learning predictive accuracy and simulated financial profitability. Furthermore, ablation studies and parameter analysis confirm that the dynamic weight generation mechanism effectively captures market reversals caused by overcrowded trades. Full article

Adenosine triphosphate (ATP) serves as a central energy currency and signaling molecule in cellular processes, with ATP-binding sites in proteins playing critical roles in enzymatic catalysis, signal transduction, and gene regulation. The accurate identification of ATP-binding sites is essential for understanding protein function mechanisms and facilitating drug discovery, enzyme engineering, and disease pathway analysis. In this study, we present a novel hybrid deep learning framework that synergizes heterogeneous learning paradigms based on protein sequence information for accurate ATP-binding site prediction. Our approach integrates two complementary base classifiers. One is a Transformer-based model, which leverages high-level contextual embeddings generated by Evolutionary Scale Modeling 2 (ESM-2), a state-of-the-art protein language model, combined with a local–global dual-attention mechanism that enables the model to simultaneously characterize short-segment and long-range contextual dependencies across the entire protein sequence. The other is a deep Q-network (DQN)-inspired classifier that achieves residue-level prediction as a sequential decision-making process. The final predictions are generated using a weighted ensemble strategy, where optimal weights are determined via cross-validations to leverage the strengths of both models. The prediction results on benchmark independent testing sets indicate that our method achieves satisfactory performance on key metrics. Beyond predictive efficacy, this work uncovers the intrinsic biological mechanisms underlying protein–ATP interactions, including the synergistic roles of local structural motifs and global conformational constraints, as well as family-specific binding patterns, endowing the research with substantial biological significance. The research in this work offers a deeper understanding of the protein–ligand recognition mechanisms and supportive efforts on large-scale functional annotations that are critical for system biology and drug target discovery. Full article

Low-dose computed tomography (LDCT) can effectively reduce ionizing radiation; however, the associated image noise and artifacts can severely compromise the accuracy of clinical diagnosis. To address the challenge of balancing noise suppression and detail preservation in LDCT images, this study proposes a deep learning (DL)-based image denoising method termed Progressive Fusion Distillation Network (PFDN). Building upon the Information Multi-distillation Network (IMDN), the proposed method incorporates a pixel attention (PA) mechanism and a progressive fusion strategy, and further designs a Pixel Parallel Extraction Block (PPEB) together with a Progressive Fusion Distillation Block (PFDB) to fully exploit multi-scale and multi-channel features, thereby optimizing the image denoising network through efficient feature separation and re-fusion. In addition, by explicitly leveraging the noise characteristics specific to LDCT images, the method establishes an end-to-end training framework suitable for medical imaging. Experimental results demonstrate that PFDN not only effectively reduces image noise and artifacts, but also enhances overall image quality while preserving diagnostically relevant image structures under the adopted evaluation setting. Full article

Wind turbine blade surface defect detection remains challenging due to large variations in defect scales, blurred edge textures, and severe interference from complex backgrounds, which often lead to insufficient detection accuracy and high false and missed detection rates. To address these issues, this paper proposes an improved RTDETR-based detection framework, termed CEA-DETR, for wind turbine blade surface defect inspection. First, a Cross-Scale Multi-Edge feature Extraction (CSME) backbone is designed by integrating multi-scale pooling and edge-enhancement units with a dual-domain feature selection mechanism, enabling effective extraction of fine-grained texture and edge features across different scales. Second, an Efficient Multi-Scale Feature Fusion Network (EMSFFN) is constructed to facilitate deep cross-level feature interaction through adaptive weighted fusion and multi-scale convolutional structures, thereby enhancing the representation of multi-scale defects. Furthermore, an adaptive sparse self-attention mechanism is introduced to reconstruct the AIFI module, strengthening global dependency modeling and guiding the network to focus on critical defect regions under complex background conditions. Experimental results demonstrate that CEA-DETR achieves mAP50 and mAP50:95 of 89.4% and 68.9%, respectively, representing improvements of 3.1% and 6.5% over the RT-DETR-r18 baseline. Meanwhile, the proposed model reduces computational cost (GFLOPs) by 20.1% and parameter count by 8.1%. These advantages make CEA-DETR more suitable for deployment on resource-constrained unmanned aerial vehicles (UAVs), enabling efficient and real-time autonomous inspection of wind turbine blades. Full article

Bone marrow cell classification and nucleus segmentation in microscopic images are fundamental tasks for computer-aided diagnosis of lymphocytic leukemia. However, bone marrow cells from different subtypes exhibit high morphological similarity, and structural information is often constrained under optical microscopic imaging, posing challenges for stable and effective feature representation. To address this issue, we propose MTLL (Multitask Model on Lymphocytic Leukemia), a novel multitask approach that performs cell classification and nucleus segmentation within a unified network to exploit their complementary information. The model constructs a hybrid backbone for shared feature representation based on a CNN-Transformer architecture, in which Fuse-MBConv modules are tightly integrated with multilayer multi-scale transformers to enable deep fusion of local texture and global semantic information. For the segmentation branch, we design an AM (Atrous Multilayer Perceptron) decoder that combines atrous spatial pyramid pooling with multilayer perceptrons to fuse multi-scale information and accurately delineate nucleus boundaries. The classification branch incorporates prior knowledge of cell nuclei structures to capture subtle variations in cellular morphology and texture, thereby enhancing the model’s ability to distinguish between leukemia subtypes. Experimental results demonstrate that the MTLL model significantly outperforms existing advanced single-task and multi-task models in both lymphocytic leukemia classification and cell nucleus segmentation. These results validate the effectiveness of the multi-task feature-sharing strategy for lymphocytic leukemia diagnosis using bone marrow microscopic images. Full article

In ecologically fragile and geomorphologically complex mountainous regions, ensuring a smooth transition from poverty alleviation to multidimensional sustainable rural development remains a key issue in regional governance. Focusing on the Qinba Mountains, a typical former contiguous poverty-stricken region in China covering 18 prefecture-level cities in six provinces, this study uses 2009–2023 prefecture-level panel data to examine the spatiotemporal evolution and driving mechanisms of coordinated rural revitalization. An integrated framework of “multi-dimensional evaluation–spatiotemporal tracking–attribution diagnosis” is developed by combining the improved AHP–entropy-weight TOPSIS method, the Coupling Coordination Degree (CCD) model, spatial Markov chains, spatial autocorrelation, and the Geodetector. The results show pronounced subsystem asynchrony. Livelihood and Well-being Security (U5) improves steadily, while Level of Industrial Development (U1), Civic Virtues and Cultural Vibrancy (U3), and Rural Governance (U4) also rise but with clear spatial differentiation; by contrast, Quality of Human Settlements (U2) fluctuates in stages under ecological fragility. Overall, the coupling coordination level advances from the Verge of Imbalance to Intermediate Coordination, yet the regional pattern remains uneven, with eastern basin cities leading and western deep mountainous cities lagging. State transitions display both policy responsiveness and path dependence: the probability of retaining the original state ranges from 50.0% to 90.5%; low-level neighborhoods reduce the upward transition probability to 25%, whereas medium-to-high-level neighborhoods raise the upward transition probability of low-level cities from 36.36% to 53.33%. Spatial dependence is also evident, with Global Moran’s I increasing, with fluctuations, from 0.331 in 2009 to 0.536 in 2023; high-value clusters extend along the Guanzhong Plain–Han River Valley corridor, while low-value clusters remain relatively locked in mountainous border areas. Driving mechanisms show clear stage-wise succession. At the single-factor level, the explanatory power of Road Network Density (F6) declines from 0.639 to 0.287, whereas Terrain Relief Amplitude (F1) becomes the dominant background constraint in the later stage (q = 0.772). Multi-factor interactions are generally enhanced. In particular, the traditional infrastructure-led pathway weakens markedly, with F1 ∩ F6 = 0.055 in 2023, while the interaction between terrain and consumer market vitality becomes dominant, with F1 ∩ F7 = 0.987 in 2023. On this basis, three major pathways are identified: government fiscal intervention and transportation accessibility improvement, capital agglomeration and market demand stimulation, and human–earth system adaptation and ecological value realization. These findings provide quantitative evidence for breaking spatial lock-in and improving cross-regional resource allocation in ecologically constrained mountainous regions. Full article

Oxidative stress is no longer viewed as a random imbalance between reactive oxygen species and antioxidants, but as a failure of an integrated redox network that connects metabolism, immunity, and metal homeostasis. Classical markers such as malondialdehyde and 4-hydroxynonenal define oxidative damage, yet they cannot explain how redox adaptation occurs or fails. Over the past decade, the discovery of regulated cell-death pathways (ferroptosis, cuproptosis) and emerging metabolic signals has revealed a new generation of adaptive redox mediators—including itaconate, nitro-fatty acids, reactive sulfur species and succinate—that act as electrophilic or persulfidating regulators rather than passive by-products of oxidation. This review integrates mechanistic, biochemical and clinical evidence to define how these mediators remodel the nuclear factor erythroid 2-related factor 2/Kelch-like ECH-associated protein 1, nuclear factor kappa-light-chain-enhancer of activated B cells, and hypoxia-inducible factor 1-alpha axes, coordinate lipid–metal–sulfur cross-talk, and shape vulnerability or resistance to ferroptosis and cuproptosis. By combining deep molecular research with translational perspectives, we propose a unifying framework for predictive redox medicine based on composite biomarker panels and AI-assisted phenotyping. Understanding and quantifying these next-generation mediators will open new avenues for precision nutrition, drug development, and disease prevention—transforming oxidative-stress biology from a descriptive field into an actionable platform for human health. Full article

To address the challenges posed by the strong uncertainty of high-proportion renewable energy sources (RES) to the secure and stable operation of distributed real-time power dispatch (D-RTPD) in new-type power systems, this paper proposes an integrated solution combining a neighborhood feature aggregation-based graph attention network (NFA-GAT) and multi-agent deep deterministic policy gradient (MADDPG). First, the D-RTPD problem is modeled as a decentralized partially observable Markov decision process (Dec-POMDP), which effectively captures the stochastic game characteristics of multi-regional agents and the partial observability of grid states. Second, the NFA-GAT is designed to enhance agents’ perception of grid operating states: by introducing a spatial discount factor, it realizes rational aggregation of multi-order neighborhood information while modeling the attenuation of electrical quantity influence with topological distance. Third, a prior-guided mechanism is integrated into the MADDPG framework to eliminate constraint-violating actions by setting their actor logits to negative infinity, improving training efficiency and strategy reliability. Simulation validations on the IEEE 118-bus test system (75.2% RES installed capacity ratio) show that the proposed method achieves efficient training convergence. Compared with the multi-layer perceptron (MLP) structure, it attains higher cumulative reward values and scenario win rates. When compared with traditional model-driven (ADMM) and data-driven (Q-MIX) methods, the proposed method balances solution efficiency, operational safety (98.7% maximum line load rate, zero power flow violation rate), and economic performance ($12,845 daily dispatch cost), providing a reliable technical support for D-RTPD under high-proportion RES integration. 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

Infrared and visible image fusion aims to integrate complementary information from heterogeneous images captured by different optical sensors based on distinct imaging principles; however, existing methods often exhibit modality bias, leading to weakened targets or the loss of crucial texture details. To address this, we propose MBFTFuse, an adversarial fusion network based on modality balancing and feature tracing, which consists of a triple-path generator and dual discriminators. The architecture employs a generator with a triple-path structure: a central modality-balancing path for deep feature fusion and dual edge feature-tracing paths for modality-specific enhancement. Specifically, a multi-cognitive modality-balancing module is introduced to achieve feature weight equilibrium, while a Feature-Tracing Attention Module self-enhances single-modality features to compensate for information loss in the fusion results. Furthermore, a pixel loss based on intensity histograms is designed to optimize inter-modal balance at the pixel level. Comparative experiments against nine state-of-the-art methods across three public datasets demonstrate that MBFTFuse effectively highlights infrared targets while preserving intricate visible textures. The superior performance of this method in both quantitative metrics and downstream object detection tasks contributes to extending the boundaries of sensor-driven computer vision technologies. 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

The Sentinel-3 Ocean and Land Colour Instrument (OLCI) is designed for water monitoring. Its 21-spectral bands serve as the basis for the precise retrieval of water quality parameters. However, its coarse resolution restricts the depiction of the spatial distribution of water quality parameters in small inland water bodies. Spatial–spectral fusion is a common method to address the inherent constraints between the spatial and spectral resolutions of sensors. Central to the popular methods is the deep learning-based method. Nonetheless, deep-learning-based models still face challenges in fusing Sentinel-2 Multi-Spectral Instrument (MSI) and Sentinel-3 OLCI data. Here, we propose a Multi-Scale-Attention-based Unsupervised Generative Adversarial Network (MSA-UGAN), which effectively integrates OLCI’s spectral advantage and MSI’s spatial resolution. Quantitative evaluation was conducted against five benchmark methods, including traditional approaches (GS, SFIM, MTF-GLP) and deep learning models (SRCNN, UCGAN). The results show that MSA-UGAN achieves the best overall performance: QNR (0.9709) and SSIM (0.9087) are the highest, while SAM (1.1331), spatial distortion (D_S = 0.0389), and spectral distortion (D_λ = 0.0252) are the lowest. This shows that MSA-UGAN can better preserve the spatial details of S2 MSI and the spectral features of S3 OLCI data. Moreover, ERGAS (2.2734) also performs excellently in the comparative experiments. The experiment of Chlorophyll-a inversion using the fused image in Chen Lake revealed a spatial gradient ranging from 3.25 to 19.33 µg/L, with the highest concentrations in the southwestern nearshore waters, likely associated with aquaculture. These results jointly indicate that MSA-UGAN can generate high-spatial-resolution multispectral images, and the fused images can be effectively utilized for water quality monitoring, thereby providing essential data support for the precision management and scientific decision-making regarding inland lakes. Full article

Low-count positron emission tomography (PET) is inherently affected by Poisson-dominated noise, which degrades image contrast, structural delineation, and quantitative reliability. This study systematically evaluated residual learning-based deep neural networks to investigate the influence of residual block depth on PET image restoration performance under low-count conditions. We employed a physically controlled striatum phantom, fabricated using 3D printing technology, to ensure reproducible acquisition conditions and controlled physical variability. PET images were acquired using a clinical PET/computed tomography (CT) system with list-mode acquisition. Low-count images reconstructed from short-duration acquisition were paired with high-count reference images reconstructed from extended acquisitions. We compared conventional filtering techniques, including median, Wiener, and modified median Wiener filters, with residual network (ResNet)-based models incorporating 8, 16, and 32 residual blocks. Image quality was quantitatively assessed using contrast-to-noise ratio (CNR), coefficient of variation (COV), line profile analysis, universal quality index (UQI), and perceptual image patch similarity (LPIPS). The results demonstrated that ResNet-based restorations substantially outperformed conventional filtering techniques in contrast recovery, signal stability, and structural preservation. The ResNet-16 model achieved the most balanced performance, yielding the highest CNR (9.02) and lowest COV (0.105), while also demonstrating superior structural and perceptual similarity, as indicated by UQI (0.9224) and LPIPS (0.0174), relative to the high-count reference images. Deeper network configurations exhibited diminishing returns and reduced structural consistencies. These findings indicate that an intermediate residual block depth is optimal for low-count PET image restoration and highlight the importance of architectural optimization in deep learning-based PET image enhancement with phantom-based evaluation frameworks. Full article

The increasing proportion of renewable energy leads to frequent changes in turbine load, making the regulating stage more prone to degradation. Traditional anomaly detection methods lack sufficient sensitivity and generalization. To address this issue, this study proposes a method combining multi-attention gated fusion and physical information learning. A gated fusion mechanism is proposed to adaptively extract and fuse key temporal and feature information. Furthermore, the generalization ability of the model is improved by introducing physical constraints derived from the relationship between pressure, temperature, and valve position. Finally, a dynamic temperature prediction model is established using the multi-output long short-term memory neural network. Experiments using actual power plant data demonstrate that the proposed method effectively improves the accuracy of post-regulating-stage temperature prediction and the sensitivity of anomaly detection. The proposed gating fusion method improves prediction accuracy by 4.6% compared to direct addition, while the fusion of physical information reduces the generalization error by more than 6%. In addition, compared to traditional deep learning and machine learning models, the proposed method improves anomaly detection accuracy by at least 3.9%. This research is of great significance for the safe operation of thermal power units and the power grid. Full article