Search Results (1,307)

Accurate photovoltaic (PV) power forecasting is essential for the reliable integration of renewable energy into electrical grids. This paper proposes a novel Multi-Scale CNN-Transformer network with Residual Correction (MSCT-RCM) for ultra-short-term PV power forecasting. The model integrates parallel multi-scale convolutional neural networks (CNNs) to extract local temporal features, a Transformer encoder to capture long-range dependencies, and a Residual Correction Module (RCM) that dynamically refines predictions using historical error patterns. A two-stage training strategy is employed to stabilize learning and enhance performance. Experimental evaluation on two years of operational data from a large-scale PV plant demonstrates that the proposed model achieves an R² value of 0.9944 for 15-minute-ahead forecasts and reduces mean absolute error (MAE) and root mean square error (RMSE) by over 50% in one-hour-ahead predictions compared to benchmark models. The MSCT-RCM model therefore exhibits strong potential for deployment in scenarios requiring high-precision predictions, such as smart grid scheduling. Full article

Background: Haemoglobinopathies are the most common monogenic disorders both in Greece and worldwide. The most effective strategies against them are carrier detection and prenatal testing following genetic risk assessment consultation for couples on the likelihood of their offspring being affected. Case Presentation: A novel alpha globin chain variant, named Hb Thessaloniki, was detected in Northern Greece. The underlying point variation HBA1:c.260T>C (ref. seq. NM_000558.5) was detected in the HBA1 gene, in heterozygosity, during a routinely performed population screening for haemoglobinopathies. The amino-acid residue Leu86 was replaced by a structure disrupting Pro residue, resulting in a hyperunstable product as shown by the isopropanol test and predicted by the Dynamut2 and Alphafold3 algorithms. The haematological phenotype, due to which genetic analysis was performed, presented with mild microcytosis and hypochromia and was also indicative of the presence of an unstable haemoglobin produced in small quantities (variant encoded by HBA1). Since the proband’s partner presented with a normal haematological phenotype, there is no risk of the couple giving birth to an affected offspring. Expanded analysis of the proband’s relatives identified biallelic variants (α^Parmaα/αα^{Τhessaloniki}) in the proband’s mother, who presented with no apparent clinical findings, expect for slightly reduced haematological indices. Conclusions: The novel Hb Thessaloniki identified, although theoretically hyperunstable, seems to have minor effects on erythrocyte function, as indicated by haematological findings on the proband and his close relatives. Future identification of co-inheritance with HBA pathogenic point variations or deletions may provide further information regarding genetic counselling. In parallel, the usage of structure–function relation-calculating algorithms may enhance our prediction capability for novel variants. Full article

User behavior anomaly detection plays a vital role in network security for identifying malicious access and abnormal activities in high-dimensional internet user behavior data. Although Transformer architectures have been widely adopted in anomaly detection tasks, and their integration with Variational Autoencoders (VAEs) has often been used to further improve detection accuracy, existing integration methods have failed to effectively balance global feature dependency modeling and generative data distribution learning. This results in limited capability in identifying complex anomalous patterns. To address this issue, this paper proposes DF-TransVAE, a novel deeply integrated framework that advances the integration of a Transformer and a VAE for supervised anomaly detection. The framework first fuses global contextual representations from the Transformer encoder with original input features, then maps the fused representation into the latent space via the VAE encoder. A cross-attention mechanism is introduced as the core of deep integration, enabling dynamic, bidirectional interaction between the fused features and latent variables to enhance information fusion. Lastly, a fully connected classifier equipped with residual connections outputs anomaly probabilities for supervised binary classification. Experimental results on two public datasets demonstrate that the proposed framework achieves better performance than existing deep learning methods in terms of accuracy, precision, recall, and F1-score, particularly in detecting complex anomalous patterns. Our results indicate that the deep integration mechanism we propose effectively addresses the limitations of conventional Transformer–VAE combinations. Full article

PET biodegradation remains limited due to its intrinsic properties—high crystallinity, hydrophobicity, and strong chemical stability. These characteristics lead to extremely slow degradation rates and contribute to PET’s persistence in the environment. Understanding how microorganisms respond at the molecular level when exposed to such a recalcitrant polymer is therefore essential. Living organisms express genes in response to their needs during development. When microbes are under critical conditions, such as when contaminants are present, they express genes encoding specific enzymes that attack the pollutant. In this study, a fungus isolated from the infected fruit of the plant Randia monantha was identified as Aspergillus terreus. It was tested for polyethylene terephthalate (PET) degradation, and the fungus Aspergillus nidulans was evaluated due to its previously reported recombinant cutinases for PET degradation. A microplastic polyethylene terephthalate (PET-MP) particle size of <355 μm for degradation was established, and a PET weight loss of 1.62% for A. nidulans and 1.01% for A. terreus was found. Additionally, the degradation of PET was confirmed by FTIR and SEM. This study also compares the transcriptomic profiles of Aspergillus nidulans and Aspergillus terreus during cultivation with PET-MP residues, which serve as a replacement for the carbon source. We present the first evidence of chitinase overexpression during direct exposure of PET to Aspergillus fungi. Interestingly, chitinase expression was detected in the crude extracts of A. nidulans and A. terreus during culture in the presence of PET residues, which replaced the carbon source. The chitinase produced by each fungus has a similar molecular weight of approximately 44 kDa. Chitinase activity was monitored over a 14-day cultivation period; from day 2, chitinase activity was detected in both cultures and continued to increase until day 14, when the highest values reported in this work were 24.88 ± 4.17 U mg⁻¹ and 10.41 ± 0.47 U mg⁻¹ for A. nidulans and A. terreus, respectively. Finally, we proposed a pathway for PET degradation by Aspergillus fungi that involves mycelial adherence and the secretion of hydrophobins, followed by the production of intermediates and monomers via esterase hydrolysis, and ultimately, the entry of monomers to the ethylene glycol (EG) and terephthalic acid (TPA) pathways, further suggesting these Aspergillus as candidates to produce valuable compounds under these conditions, such as muconic acid, gallic acid, and vanillic acid. Full article

Small unmanned aerial vehicle (UAV) detection in low-altitude infrared imagery remains challenging due to extremely small targets, weak contrast, scarce appearance cues, and heavy background clutter, which often leads to missed detections, clutter-induced false alarms, and localisation drift. To address these issues, we propose Directional Context Modelling DETR (DCM-DETR), an end-to-end detector that strengthens weak target evidence via directional context modelling and scale-consistent feature aggregation. Specifically, we build a Directional Receptive-Field Enhancement (DRFE) backbone with C2f-APC units, introducing asymmetric padding to enlarge receptive fields while preserving faint target cues. We further design an Infrared-Enhanced Encoder (IEE), where a CSA-Block jointly captures directional context and local details to steer global interactions towards target-relevant regions. To suppress noise propagation and alleviate cross-scale misalignment, we employ Hierarchical Gated Fusion (HGF) and Residual Alignment (RA), enabling selective semantic modulation and consistent multi-scale alignment. Moreover, we incorporate a Magnitude-Aware Linear Attention AIFI (MALA-AIFI) module to enhance low-SNR responses with linear complexity. Experiments on SIRST-UAVB show that DCM-DETR improves $mA{P}_{50}$ by 36.58% over YOLOv8n and by 1.09% over RT-DETR, while reducing parameters by 25.1M. On IRSTD, it yields a 2.01% gain in $mA{P}_{50}$ and boosts speed from 47.43 FPS to 93.45 FPS. These results demonstrate that DCM-DETR achieves a strong accuracy–efficiency trade-off for infrared small UAV detection in cluttered low-altitude scenes. Full article

To investigate the dynamic changes in driver attention in complex road traffic scenarios, this paper proposes a driver attention prediction method based on cross-modal adaptive feature fusion (DAFNet). First, semantic segmentation is applied to the input image sequences, and a dual-branch encoder using a 3D residual network is designed to extract spatio-temporal features from both RGB images and semantic information in parallel. Next, a 3D deformable attention mechanism is introduced to enhance the traditional Transformer algorithm, which focuses on the key salient regions through spatio-temporal offset prediction and adaptive fusion of cross-modal features. Subsequently, a predictive recurrent neural network is employed to forecast the fused spatio-temporal features and improve the stability of long-term sequence prediction. Finally, the driver attention results are predicted by a lightweight decoder. Experimental results demonstrate that the proposed method outperforms the comparative methods in overall performance. The predictions not only capture salient regions in driving scenes in a bottom-up manner but also track the driver’s intent in a top-down manner. Thus, our method exhibits strong adaptability to various complex traffic scenarios. Additionally, the method achieves an inference speed of 53.73 frames per second, satisfying the real-time performance requirement of on-vehicle systems. Full article

The application of vibration monitoring integrated with sparse Auto-Encoder (SAE) networks is investigated in this paper with the objective of detecting and localizing structural anomalies or damages. Unlike previous studies on SAE networks, the methodology proposed is based on the definition of a single SAE model, trained with the signals simultaneously collected from several sensors. Once the SAE has been trained using measurements that represent the baseline (undamaged) condition of the structure, the network is likely to reconstruct well newly collected data if the structure maintains its intact condition. When damage or structural degradation processes start developing, an increase in the reconstruction error—defined as the residual between the original input and the reconstructed output—has to be expected, so that a deviation from the normal state is highlighted. Moreover, this rise in reconstruction errors is typically more significant near the damaged areas, allowing for precise localization of the affected zones. The performance and robustness of the proposed approach are illustrated and validated using experimental data from two real-world bridge structures. Full article

Traffic flow prediction is a pivotal task in intelligent transportation systems. The primary challenge lies in accurately modeling the dynamically evolving and directional spatio-temporal dependencies inherent in road networks. Existing graph neural network-based methods suffer from three main limitations: (1) symmetric adjacency matrices fail to capture the causal propagation of traffic flow from upstream to downstream; (2) the serial combination of graph and temporal convolutions lacks an explicit modeling of joint spatio-temporal state transition probabilities; (3) the inherent low-pass filtering property of temporal convolutional networks tends to smooth high-frequency abrupt signals, thereby weakening responsiveness to sudden events. To address these issues, this paper proposes a causal-enhanced spatio-temporal Markov graph convolutional network (CSHGCN). At the spatial modeling level, we construct an asymmetric causal adjacency matrix by decoupling source and target node embeddings to learn directional traffic flow influences. At the spatio-temporal joint modeling level, we design a spatio-temporal Markov transition module (STMTM) based on spatio-temporal Markov chain theory, which explicitly learns conditional transition patterns through temporal dependency encoders, spatial dependency encoders, and a joint transition network. At the temporal modeling level, we introduce differential feature enhancement and high-frequency residual compensation mechanisms to preserve key abrupt change information through frequency-domain complementarity. Experiments on four datasets—PEMS03, PEMS04, PEMS07, and PEMS08—demonstrate that CSHGCN outperforms existing baselines in terms of MAE, RMSE, and MAPE, with ablation studies validating the effectiveness of each module. Full article

Real-time lithology identification during drilling faces challenges such as indistinct boundaries and difficulties in feature extraction. To address these, this study proposes the MSC-Transformer, a novel model integrating time-frequency features with a deep neural network. A series of drilling experiments were conducted using an intelligent drilling platform, during which triaxial vibration signals were collected from five types of rock specimens: anthracite, granite, bituminous coal, sandstone, and shale. Short-time Fourier Transform (STFT) was applied to generate multi-channel power spectral density (PSD) maps, which were then fused into a three-channel tensor to preserve directional frequency information and used as inputs to the model. The proposed MSC-Transformer combines a multi-scale convolutional (MSC) module with a lightweight Transformer encoder to jointly capture local texture patterns and global dependency features, thereby enabling accurate classification of complex lithologies. Experimental results demonstrate that the model achieves an average accuracy of 98.21 ± 0.49% on the test set, outperforming convolutional neural networks (CNNs), visual geometry group (VGG), residual network (ResNet), and bidirectional long short-term memory (Bi-LSTM) by 5.93 ± 0.90%, 2.54 ± 1.11%, 6.38 ± 2.63%, and 10.56 ± 3.11%, respectively, with statistically significant improvements (p < 0.05). Ablation studies and visualization analyses further validate the effectiveness and interpretability of the model architecture. These findings indicate that lithology recognition based on time-frequency representations of vibration signals is both stable and generalizable, offering technical support for real-time intelligent lithology identification during drilling operations. Full article

Speech-driven face generation aims to synthesize a face image that matches a speaker’s identity from speech alone. However, existing methods typically trade identity fidelity for visual quality and rely on large end-to-end generators that are difficult to train and tune. We propose Vox2Face, a speech-driven face generation framework centered on an explicit identity space rather than direct speech-to-image mapping. A pretrained speaker encoder first extracts speech embeddings, which are distilled and metric-aligned to the ArcFace hyperspherical identity space, transforming cross-modal regression into a geometrically interpretable speech-to-identity alignment problem. On this unified identity representation, we reused an identity-conditioned diffusion model as the generative backbone and synthesized diverse, high-resolution faces in the Stable Diffusion latent space. To better exploit this prior, we introduce a discriminator-free diffusion self-consistency loss that treats denoising residuals as an implicit critique of speech-predicted identity embeddings and updates only the speech-to-identity mapping and lightweight LoRA adapters, encouraging speech-derived identities to lie on the high-probability identity manifold of the diffusion model. Experiments on the HQ-VoxCeleb dataset show that Vox2Face improves the ArcFace cosine similarity from 0.295 to 0.322, boosts R@10 retrieval accuracy from 29.8% to 32.1%, and raises the VGGFace Score from 18.82 to 23.21 over a strong diffusion baseline. These results indicate that aligning speech to a unified identity space and reusing a strong identity-conditioned diffusion prior is an effective method to jointly improve identity fidelity and visual quality. Full article

As modern software systems continue to grow in scale and structural complexity, log anomaly detection has become an essential component of system monitoring and fault diagnosis. However, existing approaches often struggle to adequately capture sequential dependencies in log data and to remain robust under distributional changes. To mitigate these issues, this paper presents BERT-LogAnom, an unsupervised framework for log anomaly detection that combines contextual representation learning, sequential modeling, and adaptive decision mechanisms. Specifically, a BERT-based encoder is employed to learn global contextual semantics from log sequences, while a gated residual bidirectional Long Short-Term Memory (GR-BiLSTM) network is introduced to model bidirectional temporal dependencies without disrupting the learned contextual information. To characterize normal system behavior from unlabeled logs, two self-supervised objectives—masked log key prediction and volume hypersphere minimization—are jointly optimized during training. Furthermore, a Dynamic Thresholding Prediction Module (DTPM) is incorporated to adjust anomaly decision boundaries in response to short-term statistical fluctuations and longer-term distribution drift. Experiments conducted on three public benchmark datasets (HDFS, BGL, and Thunderbird) show that BERT-LogAnom achieves consistently superior performance compared with representative baseline methods across precision, recall, and F1-score. Additional ablation studies further confirm the contribution of each major component in the proposed framework. Full article

Molecular mechanisms determining pathogenicity of the Dutch elm disease fungus Ophiostoma novo-ulmi are poorly understood. Prior identification of the pathogenicity locus pat1 prompted a chromosome walking approach to elucidate gene function in this region. Among 17 identified genes, ONUg0282 (amtA) was predicted to encode a high-affinity ammonium transporter. In silico analyses confirmed the presence of four additional amt genes (amtB, amtC, amtD, and amtE) in both O. novo-ulmi and the less aggressive O. ulmi and that amtA and amtB belong to the Saccharomyces cerevisiae mep2 clade. The predicted amtA gene product showed features of Mep2-type transceptors, including amino acid residues corresponding to His-168 and His-318 in Escherichia coli AmtB protein, 11 transmembrane helices, and a conserved 22 amino acid motif immediately downstream of the last transmembrane helix. A knockdown amtA mutant with 25% residual expression was significantly less aggressive than wild-type O. novo-ulmi strain H327 when infecting Ulmus americana × U. parvifolia saplings. Predicted AmtA transporters from two CRISPR-Cas9 knockout mutants contained only five intact transmembrane helices. The ΔamtA mutants retained several wild-type phenotypic traits, including yeast–mycelium dimorphism, but were significantly less aggressive than H327 towards U. americana saplings. We concluded that ONUg0282 is an important determinant of aggressiveness in O. novo-ulmi. Full article

Adverse weather removal aims to restore images degraded by haze, rain, or snow. However, existing unified models often rely on implicit degradation cues, making them vulnerable to inaccurate weather perception and insufficient semantic guidance, which leads to over-smoothing or residual artifacts in real scenes. In this work, we propose AWR-VIP, a prior-guided adverse weather removal framework that explicitly extracts semantic and perceptual priors using a frozen vision–language model (VLM). Given a degraded input, we first employ a degradation-aware prompt extractor to produce a compact set of semantic tags describing key objects and regions, and simultaneously perform weather-type perception by prompting the VLM with explicit weather definitions. Conditioned on the predicted weather type and selected tags, the VLM further generates two levels of restoration guidance: a global instruction that summarizes image-level enhancement goals (e.g., visibility/contrast) and local instructions that specify tag-aware refinement cues (e.g., recover textures for specific regions). These textual outputs are encoded by a text encoder into a pair of priors (${\mathbf{P}}_{global}$ and ${\mathbf{P}}_{local}$), which are injected into a UNet-based restorer through global-prior-modulated normalization and instruction-guided attention, enabling weather-adaptive and content-aware restoration. Extensive experiments on a combined benchmark show that AWR-VIP consistently outperforms state-of-the-art methods. Moreover, the VLM-derived priors are plug-and-play and can be integrated into other restoration backbones to further improve performance. Full article

Ships target detection plays a crucial role in safeguarding maritime transportation. However, affected by factors such as ocean waves, extreme weather, and target diversity (e.g., large size differences, arbitrary rotation, and occlusion), existing deep learning-based detection methods struggle to achieve a satisfactory balance among accuracy, speed, and model size in complex marine environments. To address this challenge, this paper proposes a real-time ship detection algorithm (C-YOLO) integrating global perception and multi-scale feature enhancement. First, a Transformer encoder is added before the detection head, which suppresses interference from sea clutter and cloud mist occlusion through long-range dependency modeling, improving the detection of small and occluded ships. Second, a Dual-Effect Focused Residual Fusion Module is designed to replace the backbone’s multi-scale pooling structure, combining the advantages of CBAM (background noise suppression) and SK-Net (dynamic scale adaptation) to simultaneously capture features of ships of different sizes. Finally, a CZIoU loss function is proposed, which integrates constraints on angle, center point, vertex, and area to address rotation, deformation, and multi-scale issues in ship detection. Experimental results on the SeaShips 7000 dataset show that the proposed C-YOLO achieves a Recall of 0.842, mAP@50 of 0.797, and mAP@50:95 of 0.552, outperforming mainstream algorithms such as YOLOv7 (Recall = 0.785, mAP@50 = 0.781), YOLOv9s (Recall = 0.819, mAP@50 = 0.755), and SSD (Recall = 0.802, mAP@50 = 0.833). With 76.75 M parameters and an inference speed of 119 FPS, the model maintains efficient real-time performance while ensuring detection accuracy. This method effectively reduces false detection and missed detection rates in complex scenarios such as port monitoring and maritime traffic control, providing a reliable technical solution for intelligent maritime surveillance and safe navigation—with significant practical value for improving maritime transportation efficiency and reducing safety risks. Full article

Multi-agent trajectory prediction is essential for autonomous driving systems, as its performance heavily depends on the quality of feature representations. This paper proposes PFR-HiVT, a lightweight and effective approach for multi-agent trajectory prediction, and evaluates it on the Argoverse 1.1 motion forecasting dataset. Although existing methods such as the Hierarchical Vector Transformer (HiVT) have achieved strong performance, they still exhibit limitations in feature extraction and feature transition across different stages of the network. To address these limitations, a collaborative feature enhancement framework is introduced, consisting of two encoder-side modules and a Progressive Feature Refinement Global Interactor (PFR-Global Interactor). Specifically, the Feature Enhancement Module (FEM) and the Attention Enhancement Module (AEM) are employed to refine local spatiotemporal features before global interaction. In addition, the PFR-Global Interactor integrates three lightweight components—the Simple Feature Refinement Module (SFR), the Lightweight Gate Module (LG), and the Residual Connection Module (RC)—to progressively refine globally interacted features prior to trajectory decoding. All proposed modules adopt lightweight designs, introducing only 230.5 k additional parameters (approximately 8.7% of the total parameters of HiVT-128). Experiments on the Argoverse 1.1 dataset show that PFR-HiVT achieves a minADE of 0.703, a minFDE of 1.041, and an MR of 0.112, outperforming the baseline HiVT model. Ablation studies further validate the effectiveness and synergy of the proposed modules. Full article