Search Results (3,536)

Magnetic resonance imaging (MRI) remains one of the most important tests for diagnosing and monitoring various diseases. In recent years, machine learning methods have been widely applied to automate MRI analysis. It supports decision-making by predicting disease and highlighting relevant regions. However, the proper use of feature extraction methods can improve the performance of the model. This paper proposes a WaveletFusion architecture that combines a two-dimensional Haar wavelet decomposition with a convolutional neural network (CNN) for classification. The approach was demonstrated on the Brain Tumor MRI dataset and further examined on the Br35H :: Brain Tumor Detection 2020 (Br35H). The model decomposes each MRI slice into approximation and directional detail subbands and fuses multi-scale wavelet features within the convolutional pipeline. To evaluate the effect of decomposition depth, WaveletFusion variants from one to eight levels were compared with a Baseline CNN model under the same training protocol. The results showed that performance improved progressively with increasing decomposition depth up to level 7, whereas the 8-level configuration consistently declined, indicating that excessive decomposition introduces information loss and over-compression in the deepest approximation pathway. The best-performing configuration, which outperformed both the Baseline CNN and the WaveletFusion variations in five independent runs, was the 7-level WaveletFusion model, achieving a test accuracy of 0.94 ± 0.01 and test macro-F1 of 0.93 ± 0.02. A similar tendency was observed on the Br35H dataset, where the 7-level model achieved a 0.97 ± 0.01 test accuracy and 0.97 ± 0.01 test macro-F1, while the 8-level configuration remained weaker on both datasets. These results show that multi-scale wavelet fusion can improve Brain Tumor MRI classification while maintaining a compact model size and a fair comparison setting, and that the decomposition depth must be selected carefully. Full article

The rising incidence of skin cancer necessitates scalable and accurate diagnostic tools. While dermoscopy-based systems have achieved expert-level performance, clinical smartphone images pose challenges due to variability in lighting, resolution, and artifacts. Recent advances in multimodal deep learning have shown promise, yet most approaches rely on simple feature concatenation or single-model classifiers, limiting their ability to capture complex cross-modal interactions. This study aims to bridge the diagnostic gap in resource-limited settings by developing a robust multimodal framework that synergizes clinical smartphone images with structured patient metadata for automated skin cancer classification. We propose a novel hybrid architecture integrating a Swin Transformer V2 (SwinV2-Tiny) for hierarchical visual feature extraction and a Denoising Autoencoder (DAE) with PCA for robust metadata embedding. These heterogeneous modalities are fused via a Gated Attention Mechanism that dynamically weighs feature importance across streams. Classification is performed by a Heterogeneous Meta-Stack Ensemble comprising CatBoost, LightGBM, XGBoost, and Logistic Regression, designed to maximize calibration and generalization across imbalanced classes. Evaluated on the PAD-UFES-20 dataset (2298 clinical smartphone images, six diagnostic classes), the proposed framework achieves state-of-the-art performance with a macro-averaged F1-score of 0.977, accuracy of 0.978, and an AUC of 0.990. It significantly outperforms unimodal baselines and existing multimodal methods, demonstrating superior sensitivity (0.974) and precision (0.981), particularly for underrepresented malignant classes like Melanoma (F1: 0.995) and Squamous Cell Carcinoma (F1: 0.960). The integration of clinical metadata with advanced visual embeddings via gated attention significantly enhances diagnostic reliability. Comprehensive ablation studies confirm the contribution of each architectural component. This framework offers a viable pathway for deploying high-precision, AI-driven dermatological screening tools on standard smartphone devices. Full article

Adhesion between fused deposition modelling (FDM) printed polymers and textile substrates is critical for durable printed-on-textile hybrids. Since no dedicated test standard exists for additively manufactured textile interfaces, many studies use T-peel methods adapted from adhesive-bond standards. However, printed-on-textile joints are often governed by polymer penetration into the fabric and mechanical interlocking, rather than by a discrete adhesive layer. This work evaluates a fixture-based perpendicular (normal-separation) tensile method, using a circular dolly printed directly onto a cotton plain-weave substrate and a fully 3D-printable, threaded, self-aligning clamping assembly. Three representative filaments, namely polyethylene terephthalate glycol-modified (PETG), polylactic acid (PLA), and thermoplastic polyurethane (TPU), were tested using both the proposed pull-off method and an ISO 11339-type T-peel benchmark, with n = 8 specimens per polymer. The perpendicular method produced complete datasets for all polymers and clearly differentiated adhesion performance (TPU > PLA > PETG). In contrast, for T-peel, the standard evaluation window (25–125 mm) was completed for all PETG specimens but only for a subset of PLA specimens and a single TPU specimen. In the remaining tests, premature substrate failure prevented completion of this window, so the results could not be evaluated. Microscopy confirmed distinct interlocking morphologies across polymers, supporting the observed differences in failure behavior between peel and normal separation. Overall, the results indicate that perpendicular dolly pull-off testing is a practical and reproducible alternative for quantifying adhesion across a wider range of printed-on-textile bonding conditions. Full article

Instance segmentation accurately extracts the position and outline of ships, serving as the foundation for maritime safety tasks such as multi-object tracking, sensor fusion, and collision warning. This study focuses on single-frame segmentation and aims to address the challenge of multi-scale ship occlusion in congested ports, providing reliable observational data through high-precision recognition to ensure navigation safety. Existing methods suffer from performance degradation in complex maritime environments due to factors such as multi-scale distribution, low resolution of distant targets, and frequent occlusions. Among these, ship occlusion is particularly problematic as it leads to feature confusion between adjacent instances and inaccurate boundary segmentation. To address these challenges, we propose a novel instance segmentation algorithm (MAF-ORDNet) based on Multi-branch Adaptive Feature Fusion and Occluded Region Decoupling. Firstly, a multi-branch adaptive feature fusion module is designed to capture contextual information through different receptive fields and dynamically fuse multi-scale features, thereby restoring occluded semantics and enhancing robustness. Secondly, an occlusion region decoupling module is constructed to accurately localize occluded regions and enhance contour responses via adaptive sampling, achieving refined boundary processing. In addition, we constructed and annotated the Occlusion ShipSeg dataset, which contains 1969 real occlusion images, 2150 simulated occlusion images, and 1132 images under adverse weather conditions, totaling 17,352 fine instance annotations. Experimental results show that, compared with PatchDCT, YOLOv11s, and Mask2Former, our method improves AP by 2.7%, 3.2%, and 2.4%, respectively, while maintaining a comparable inference speed to YOLOv8s. These results confirm that MAF-ORDNet achieves a favorable balance between accuracy and efficiency in multi-scale occluded ship segmentation tasks. Full article

Landslides are widespread geohazards in mountainous regions and pose serious threats to human safety, infrastructure, and ecosystems. Accurate detection from high-resolution optical remote sensing imagery remains challenging because landslide targets often exhibit irregular morphology, large scale variation, weak boundaries, and strong background interference. To address these issues, this study proposes L-SAINet, a shape-adaptive and inner-scale interaction network for landslide detection in complex remote sensing scenarios. Built on a lightweight one-stage detection framework, the proposed method introduces an L-SAI module that integrates adaptive deformable convolution, channel–spatial attention, and inner-scale feature interaction. The shape-adaptive branch improves geometric alignment for irregular and elongated landslide bodies, while the attention branch enhances semantic discrimination under heterogeneous background conditions. The two branches are further fused at the same feature scale to construct a more unified landslide representation. Experiments on the Bijie Landslide Remote Sensing Dataset show that L-SAINet consistently outperforms the baseline detector and single-branch variants in Precision, Recall, mAP@0.5, and mAP@0.5:0.95. Additional analyses based on precision–recall curves, confusion matrices, convergence behavior, model complexity, and representative complex-scene examples further confirm its effectiveness and robustness. The results demonstrate that jointly modeling geometric adaptability and semantic refinement is an effective strategy for landslide detection in complex mountain environments. Full article

Multi-label image classification (MLIC) is vulnerable to contextual bias, where models may exploit spurious label–context associations rather than object evidence, leading to degraded generalization under distribution shifts. To address this issue, we propose CPAN, a causal-inspired framework that integrates label-specific feature decoupling, prototype-based mediator modeling, patch-level evidence aggregation, and adaptive fusion. Specifically, CPAN uses a Transformer decoder to extract label-specific representations from the whole image and local patches. We introduce a prototype dictionary as a surrogate mediator space to encourage the model to rely on object-relevant intermediate patterns rather than context-sensitive shortcuts. We further aggregate patch-level predictions to enhance direct object evidence and fuse them with whole-image predictions through a learnable gate. Experiments on two benchmark datasets show that CPAN consistently improves both recognition accuracy and robustness. On MS-COCO, CPAN achieves 85.26 mAP, 80.67 CF1, and 82.52 OF1; on NUS-WIDE, it reaches 66.11 mAP, 64.42 CF1, and 75.95 OF1. Under context-shifted evaluation on MS-COCO, CPAN further obtains 80.93 mAP, 75.84 CF1, and 77.87 OF1, indicating stronger robustness to contextual bias. These results show that CPAN learns more object-centered representations and reduces reliance on spurious contextual correlations. Full article

Existing ship trajectory clustering methods often overlook the impact of navigation behaviors (e.g., heading and speed variations) on clustering performance. To address this limitation, a novel ship trajectory clustering method that explicitly incorporates navigation behavior sequence is proposed. Firstly, ship trajectories are preprocessed, and key motion parameters, including the ship Rate Of Turn (ROT) and acceleration at each trajectory point, are calculated through a sliding window. Secondly, by integrating various motion parameters, the navigation behaviors corresponding to trajectory points are classified, and the classification results are taken as the core element to measure the behavior distance between different trajectories. Then, the spatial distance between trajectories is measured based on the Hausdorff distance. Finally, an adaptive Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm is adopted, which fuses behavior distance and spatial distance, to realize ship trajectory clustering that takes navigation behavior into account. Experimental results on Dalian Port and Yantai Port datasets show that: (1) Compared with the classical DBSCAN and Multi-dimensional Density-Based Trajectory Clustering of Applications with Noise (MD-DBTCAN) methods, the proposed method achieves finer granularity of clustering results; (2) Compared with the classical DBSCAN method, the proposed method can effectively distinguish straight-line navigation trajectories from trajectories with frequent turning behaviors; compared with the MD-DBTCAN method, the proposed method can distinguish normal straight-line navigation trajectories from trajectories with frequent acceleration and deceleration behaviors. Full article

Quicker identifying abilities are required due to the rising frequency and severity of wildfires. Although polar-orbiting satellites with medium and high resolution can accurately identify wildfires, the majority of available fire detection images originate from such platforms. However, their low temporal revisit rates restrict the potential for early warning. Geostationary satellites provide minute-level, continuous monitoring that corresponds with the quick onset of wildfires; however, their dependence on conventional threshold methods and coarse spatial resolution result in notable detection errors. This study developed an integrated deep learning framework for accurate wildfire detection in low-resolution geostationary imagery in order to get over these restrictions. A novel dynamic index, the Dynamic Normalized Burn Ratio—Thermal (DNBRT), was proposed to characterize wildfire progression by integrating instantaneous thermal anomalies with dynamic vegetation signals. Based on this, a Fire Spatiotemporal Network (FST-Net) was designed, with an efficient residual backbone, a Convolutional Block Attention Module (CBAM) for feature refinement, and a Bidirectional Long Short-Term Memory (BiLSTM) network to capture temporal evolution. Trained and evaluated on an FY-4B-based fire/non-fire dataset, the proposed framework demonstrated superior performance. FST-Net outperformed benchmark models, improving accuracy and recall by averages of 10.30% and 9.32% respectively while achieving faster inference speed. An ablation experiment confirmed the critical role of fusing thermal and vegetation features in DNBRT, with 92.7% accuracy and 94.9% recall. Compared to the FY-4B fire product, the proposed framework enables earlier detection, maintains more complete tracking of fire progression, and exhibits greater robustness under complex burning conditions while achieving sub-hectare (0.36 ha) detection sensitivity at the 2 km resolution. By synergizing a discriminative dynamic index with an efficient spatiotemporal architecture, this work provides an effective solution for operational, real-time monitoring of small and early-stage wildfires from geostationary satellites. Full article

Accurate prediction of binary stock price movements remains a challenging task due to the coexistence of short-term noise and medium-term trend dynamics in financial time series. Existing recurrent models typically encode raw price sequences within a single representation stream and aggregate temporal information using softmax-based attention, which often entangles noisy fluctuations with underlying trends and limits nonlinear expressiveness in the final classification stage. In this paper, we propose PULSE-KAN (Price-aware Unified Linear-attention and Smoothed-trend Encoder with Kolmogorov–Arnold Network Head), a modular neural architecture designed to enhance binary stock movement prediction. The proposed framework introduces three plug-and-play components designed for LSTM-based pipelines as demonstrated here within the Adv-ALSTM framework. First, the P-EMA Trend Bridge constructs an explicit smoothed trend representation via a parameterized exponential moving average and fuses it with the raw price stream to improve trend awareness. Second, the Pola Pulse Router performs efficient temporal aggregation using linear-complexity polarized attention combined with local convolutional priors, enabling better capture of multi-scale temporal dependencies. Third, the KAN Signal Refiner replaces the conventional linear prediction head with learnable Chebyshev-polynomial activations, providing enhanced nonlinear modeling capacity for decision boundaries. Extensive experiments on two public benchmark datasets demonstrate that PULSE-KAN consistently outperforms strong recurrent and attention-based baselines in terms of both classification accuracy and the Matthews Correlation Coefficient. Further ablation studies verify that each proposed component contributes independently and significantly to the overall performance improvement. Full article

To address the urgent demand for real-time reconstruction in remote sensing satellite imaging, as well as the difficulty of extracting sparse target features from dark backgrounds under low-illumination conditions, this paper proposes a lightweight, scalable attention-guided super-resolution reconstruction framework (SASR). The framework adopts an efficient, scalable visual backbone with staged feature extraction to capture discriminative information at three hierarchical scales. A refined multi-scale channel attention module, improved from the classic MS-CAM structure, is further introduced to fuse high-level semantic features and low-level texture details comprehensively. Finally, stacked sub-pixel convolution operations are employed to achieve high-precision image super-resolution enhancement. The proposed method maintains superior lightweight characteristics and fast inference efficiency while embedding effective channel attention optimisation for accurate feature representation. Experimental validations are conducted on the GF-5 satellite datasets: at 2× magnification, the proposed model achieves 32.2346 dB PSNR and 0.8791 SSIM; at 3× magnification, 31.6040 dB PSNR and 0.8376 SSIM; at 4× magnification, PSNR remains above 30 dB, and SSIM exceeds 0.8. The framework also exhibits robust generalization performance on marine remote sensing image datasets. Comparative experiments with recent super-resolution methods on multiple public datasets further verify the effectiveness and practical superiority of the proposed approach. Full article

Rapid urbanization and the rapid accumulation of multi-source and multi-temporal Earth observation data are creating an increasing demand for remote sensing models that can flexibly support fine-grained monitoring beyond fixed label taxonomies. Open-vocabulary remote sensing image semantic segmentation (OVRSIS) aims to segment text-specified categories beyond a fixed label space with vision–language foundation models. However, dense remote sensing scenes make pixel–text matching highly vulnerable to semantic confusion and misalignment, owing to extreme scale variation, thin structures, repetitive textures, and prompt sensitivity. To address these challenges, we propose UGDMoE, an uncertainty-guided mixture-of-experts framework for OVRSIS. First, we design a domain-specific MoE decoder with three geometrically specialized experts—for slender structures, mid-scale objects, and large-region context—routed by the alignment-risk cue $U_0$. Second, we introduce a lightweight prompt–response estimation strategy that quantifies prediction dispersion across semantically equivalent prompts to derive $U_0$ in an annotation-free manner. Third, we develop prompt ensemble-based likelihood calibration (PELC), which takes the shared alignment-risk cue $U_0$ as input to calibrate prompt-specific logits before refinement. Finally, we design a lightweight uncertainty-aware structure refinement module that, guided by $U_0$, selectively fuses early visual features with segmentation logits to restore boundary continuity and connectivity of thin structures. We conduct extensive experiments on eight OVRSIS benchmarks under cross-dataset evaluation protocols. Trained on DLRSD, it achieves 46.97 m-mIoU and 63.31 m-mACC, surpassing the strongest baseline by 0.76 and 0.62 points; trained on iSAID, it reaches 37.47 m-mIoU and 58.52 m-mACC, improving over the strongest competitor by 0.71 and 0.61 points. UGDMoE consistently achieves state-of-the-art performance and remains robust under training-source changes. Full article

Accurate echo classification is crucial for Millimeter-wave Cloud Radar (MMCR) data quality control. Existing approaches, however, often struggle to generalize across complex scenes or lack physical interpretability. Here we propose PhySNet, a physics-guided network that combines thermodynamic priors with spatiotemporal radar features, embedding physical information across the full pipeline from feature extraction to final outputs. Based on the coupling between the lifting condensation level (LCL) and daytime clear-air echo heights, and the lagged correlation between nocturnal clear-air echo heights and their daytime counterparts, we design a physics-constrained gating block (PCGB). The PCGB extracts thermodynamic states and evolution trends from collocated surface observations, generating a clear-air echo probability map that weights the initial radar features. Building on this, we add a parallel regression branch of effective-clutter-height (ECH). This branch fuses thermodynamic features with radar spatiotemporal features, enabling the model to learn to predict the clear-air echo boundary. Finally, we apply an adaptive height filter using the predicted ECH sequence to refine the classification results. Evaluated on a multi-region, multi-season dataset from China, PhySNet achieves a probability of detection (POD) of 98.28% for meteorological echoes and 95.87% for clear-air echoes, outperforming conventional methods. By coupling data-driven learning with physical rules, our approach provides a high-accuracy, interpretable solution for cloud radar clear-air echo identification. Full article

Hyperspectral imaging (HSI) provides rich spectral information and serves as a non-destructive technique for forensic stain analysis. Conventional approaches often exhibit degraded performance due to the high dimensionality and spectral redundancy inherent in hyperspectral data. To address this challenge, a hyperspectral dataset comprising nine beverage stains—papaya, coffee, pomegranate, orange, tea, wine, whisky, rum, and brandy—is developed. Building on this dataset, an ensemble framework that combines an optimized autoencoder (AE), channel-attention (CA)-enhanced one-dimensional convolutional neural networks (1D CNNs), and a Limited Memory Broyden–Fletcher–Goldfarb–Shanno (L-BFGS-B)-based weighted fusion strategy is proposed. The autoencoder learns compact latent representations from the 204-band hyperspectral vectors, reducing redundancy while preserving discriminative spectral features. CA emphasizes informative spectral bands and improves stain separability. Multiple 1D CNN models are trained using different latent dimensionalities, and their class probability outputs are fused through an optimized L-BFGS-B weighting scheme, where higher-performing models contribute more strongly to the final decision. Experimental results demonstrate classification accuracies of 96.54%, 97.19%, and 97.86% for the AE32 CA, AE64 CA, and AE128 CA models, respectively, with the optimized ensemble achieving an accuracy of 98.28%. Additionally, the time-dependent evolution of beverage stain reflectance is systematically analyzed using overlapped, normalized reflectance signatures acquired at time intervals of 0 min, 1 h, 2 h, 3 h, 4 h, and 5 h. The results confirm that AE-based latent compression, CA, and L-BFGS-B optimized ensemble fusion enhance hyperspectral beverage stain classification, providing an effective and extensible framework for forensic trace evidence analysis. Full article

Deep-learning-based vulnerability detection has made substantial progress, but two limitations remain prominent. Sequence-based methods linearize source code and thus weaken the explicit modeling of control-flow and data-flow dependencies. Graph-based methods preserve program structure, yet conventional graph neural networks still have difficulty capturing long-range interactions in large code property graphs (CPGs). In addition, standard CPGs usually lack explicit variable semantics and security-critical node roles, which limits their ability to represent vulnerability-relevant program behavior. To address these issues, we propose SV-GEN, a vulnerability detection framework that combines large-language-model-driven semantic enhancement with hybrid sequence-graph learning. The novelty of SV-GEN lies in introducing a semantically enriched code property graph, termed Sem-CPG, which augments conventional CPGs with variable semantic roles and security-oriented node labels, and in coupling this representation with an adaptive fusion mechanism over structural and sequential views. Specifically, we use a large language model as an external semantic annotator to assign variable roles and identify source, sink, and sanitizer nodes, and then encode the resulting Sem-CPG with a Graph Transformer while modeling the code sequence with GraphCodeBERT. A learnable gating module is further used to adaptively fuse the graph-level and sequence-level representations for final prediction. Experiments on Devign, ReVeal, and DiverseVul show that SV-GEN achieves competitive or superior overall performance across benchmarks, with particularly strong improvements on the large and highly imbalanced DiverseVul dataset. Full article

The interpretation of geophysical multi-attribute surveys is often subjective and complicated by large datasets, prompting the need for automated fusion methods that preserve structures and enhance anomalies. This study introduces an image fusion approach that combines the non-subsampled contourlet transform (NSCT) with the New Metric Parameter (NMP) rule to integrate multi-source polarizability and resistivity data for copper–nickel exploration. Using NSCT, source images are decomposed into multi-scale, multi-directional low- and high-frequency sub-bands. Low-frequency components are fused through dynamic weighting, while high-frequency components are merged using the NMP rule. The sensitivity to key parameters—such as low-frequency weight, grid size, and grid angle—was assessed using field data. Results indicate that NSCT + NMP fusion enhances spatial resolution and boundary definition of anomalies, effectively merging low resistivity with high polarizability signals. Quantitative field validation shows that 82.43% of the gabbroic mineralization zone has a judging coefficient below 0.45, confirming the fusion accuracy. Optimal parameter choices include dynamically adjusted low-frequency weights, a grid size that balances detail and noise suppression, and a 45° square grid for directional neutrality. This method offers a practical strategy for joint multi-physical data analysis and improved spatial recognition of mineralized bodies in exploration. Full article