Search Results (209)

Conservation-oriented renewal of traditional rural settlements increasingly requires evidence-based color control that considers both natural environmental backgrounds and built-cultural interfaces. This study examined whether built-cultural colors in a traditional mountain village are differentiated from natural environmental colors in hue composition while remaining proximate in NCS attribute space and explored how such quantitative findings can inform carrier-specific architectural color-control guidance. Taking Qingmuchuan Village in the Qinba Mountain region as a case study, 145 representative color samples were recorded, including 59 natural environmental samples and 86 built-cultural environmental samples. The samples were encoded using the Natural Color System (NCS) and their hue composition, blackness–whiteness–chroma attributes, nonparametric differences, exploratory structural order assessment, and attribute-space proximity were analyzed. Among the retained carrier-oriented samples, natural environmental samples were dominated by green-yellow hues (54.2%), whereas built-cultural environmental samples mainly contained yellow-red, red-blue, and neutral hues (31.4%, 18.6%, and 12.8%, respectively). Blackness did not differ significantly between the two systems, while whiteness and chroma differed significantly; the mean pairwise cosine similarity was 0.824, indicating attribute-space proximity rather than direct hue correspondence. Based on these empirical results, the study proposes provisional, carrier-specific guidance for facade renewal, roof and eave replacement, paving repair, and signage regulation. Full article

UAV path planning in complex static 3D environments involves multiple conflicting objectives and intricate constraints. However, when applied to highly constrained path planning tasks, MOWOA often suffers from a low proportion of feasible solutions, convergence instability, single-leader search bias, and an uneven distribution of Pareto solutions. To address these issues, this study formulates the UAV path planning problem as a multi-objective optimization problem that simultaneously considers path length, threat cost, smoothness cost, and altitude cost, and proposes an improved multi-leader multi-objective whale optimization algorithm (IML-MOWOA). The proposed IML-MOWOA progressively improves three key stages of the optimization process: initial population construction, search guidance, and external archive maintenance. Specifically, an adaptive opposition-based learning initialization strategy is first introduced to improve the feasibility and spatial coverage of initial paths. Based on the resulting non-dominated solution set, a grid-based external archive update strategy is then used to regulate solution density and provide representative candidate leaders from sparse Pareto regions. Subsequently, a multi-leader dynamic weighted search mechanism with Softmax-based cosine annealing integrates these leaders into the WOA update process, thereby enhancing multi-directional path exploration and alleviating premature convergence. Comparative experiments conducted in three static 3D environments of varying complexity demonstrate that the proposed method achieves more robust convergence, better Pareto-front distribution, and more balanced task-level path quality than the benchmark algorithms. In the most challenging scenario, IML-MOWOA achieves the highest number of feasible paths, reduces the mean IGD by 25.04%, and decreases the mean path length, threat cost, smoothness cost, and altitude cost by 1.65%, 28.45%, 53.23%, and 29.88%, respectively, compared with the best-performing competing algorithm for each metric. These results indicate that the proposed algorithm is effective and robust for constrained multi-objective UAV path planning in complex static 3D environments. Full article

This study proposes VulnPattern-TKG, an end-to-end temporal knowledge graph framework that forecasts the emergence of CVE-derived vulnerability-pattern relations from Common Vulnerabilities and Exposures (CVE) descriptions. The framework does not aim to predict the real-world exploitation of individual CVEs; instead, it models how standardized relations among Weakness Factor (WF), Exploitation Outcome (EO), and Exploitation Prerequisite (EP) categories evolve over time in vulnerability disclosure text. It processes 205,600 National Vulnerability Database (NVD) CVE descriptions from 2014 to 2024 using a hybrid pipeline combining SecureBERT+CRF-based entity extraction, dependency-parsing-based relation rules, and four-stage hierarchical standardization. The resulting compact Knowledge Layer contains 26 standardized category nodes and 48,371 confidence-filtered triples. VulnTEC is a lightweight confidence- and time-weighted Node2Vec graph embedding framework that ranks relation-compatible candidate tails using cosine similarity over shared node embeddings. An internal four-component priority-score framework, integrating prediction confidence, temporal rise, exploitation-prerequisite prevalence-risk proxy, and extraction confidence, supports an analyst-side review of the forecasted relations. Under the novel-only triggers evaluation, VulnTEC achieves a mean MRR of 0.410 ± 0.020; however, the theoretical random baseline already reaches 0.408 because the candidate tail space contains only six EO categories. The results are interpreted as directional ranking evidence, and query-level Top-K results are reported only as descriptive analyst-side review evidence. Full article

Sensor-based environmental monitoring systems are often affected by missing, noisy, and unreliable measurements caused by sensor faults, sparse deployment, calibration drift, and communication interruptions. To address these challenges, this study proposes an intelligent virtual sensor generation framework that integrates physical-constraint-based preprocessing, statistical virtual sensor modeling, KL-divergence-based fusion, deep generative augmentation, and temporal prediction. The raw weather-station data are first refined using threshold-based filtering, physical validity constraints, and Isolation Forest-based outlier detection. To handle the circular nature of wind direction, the angle is encoded using sine and cosine components during modeling and reconstructed using the atan2 function for evaluation. Multiple statistical methods, including Inverse Distance Weighting, Kernel Density Estimation, Ridge Regression, and Copula-based modeling, are employed to generate complementary virtual sensor data. These outputs are adaptively fused using KL divergence according to their distributional similarity with real sensor data. The fused datasets are further augmented using Variational Autoencoders and Conditional Tabular Generative Adversarial Networks, and then evaluated using BiLSTM and BiGRU models with MAE, MSE, and RMSE metrics. The experimental results demonstrate that the proposed framework generates physically valid and distributionally consistent virtual sensor data. Fusion-based methods outperform standalone approaches, while VAE-based augmentation generally provides better statistical fidelity and lower prediction errors than CTGAN. Additional validation using a public NOAA weather-station dataset further supports the transferability of the proposed fusion-based virtual sensing workflow. Comparisons with TimeGAN and diffusion-based temporal generative baselines, supported by Wilcoxon signed-rank testing, confirm the statistical significance and competitive performance of the proposed framework. A quantitative computational analysis also demonstrates the practical feasibility of the framework in terms of training time, inference time, memory consumption, and scalability. Overall, the proposed framework offers a reliable and scalable solution for virtual sensing in sensor-sparse and fault-prone environmental monitoring systems. Full article

Water injection allocation is critical for maintaining pressure support in mature offshore waterflooding reservoirs, but its optimization is complicated by delayed injection–production responses, interwell interference, limited intervention windows, and incomplete field labels for injector–producer connectivity. This study proposes a connectivity-aware optimization framework that couples an attention-based connectivity identification network, a group-level long short-term memory (LSTM) production surrogate, and particle swarm optimization (PSO). The methodological novelty lies in using prescribed connectivity labels in a field-informed semi-synthetic benchmark to quantitatively test whether dynamic injection–production sequences and static well-pair attributes can be transformed into interpretable connectivity estimates for injection allocation decision support. The benchmark contains five injectors, ten producers, daily injection and production histories, static well-pair attributes, response lags, and normalized connectivity coefficients generated under practical injection rate, lag, water cut, and adjustment constraints. The attention model recovered the dominant injector–producer relationships with MAE = 0.0146, RMSE = 0.0240, R² = 0.9835, cosine similarity = 0.9962, and top-three overlap = 100%. The group-level LSTM achieved MAE = 4.524 m³/d, RMSE = 5.963 m³/d, MAPE = 1.255%, and R² = 0.964 on the chronological test set. Across 15 optimization cases, the PSO module generated feasible injection reallocations under single-well rate, total-injection balance, and +/−15% adjustment constraints. The results should be interpreted as controlled methodological validation rather than direct field deployment; further testing with anonymized field data is required. Full article

It is well-known that activation functions are crucial in determining spectral expressiveness, training dynamics, and reconstruction accuracy in implicit neural representations (INRs), which employ coordinate-based multilayer perceptrons to represent continuous signals. Despite showing excellent performance, sinusoidal activations, for example SIREN, are limited in their adaptability to diverse signal types due to their fixed harmonic structure. In this paper, we propose two novel periodic activation functions for INRs. (1) Harmonic generalizes sinusoidal activations by combining the fundamental frequency with learned second and third harmonics through per-neuron trainable amplitude coefficients, resulting in a richer spectral basis within the SIREN initialization framework. (2) PM-FINER (Phase-Modulated FINER) extends the variable-periodic FINER activation by embedding frequency modulation synthesis directly into the instantaneous phase, enabling data-driven phase distortion via a learnable modulation index and carrier ratio. We conducted comprehensive experiments spanning nine architectural configurations (including SIREN, WIRE, FINER, Gaussian, Harmonic, PM-FINER, and an additional direct comparison against the Subtractive Modulative Network (SMN)), using six natural images, three learning rate schedulers, and three random seeds, totaling 486 main training runs (534 runs total including an ${\omega }_0$ sensitivity sweep). Our evaluation combined peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), and rigorous statistical analysis, such as paired t-tests, Wilcoxon signed-rank tests, Cohen’s d effect sizes, and Friedman rank tests. Under cosine annealing, Harmonic achieves a mean PSNR gain of 6.08 dB over SIREN and 2.57 dB over FINER (both $p<0.001$, Cohen’s $d>3.7$), while PM-FINER ranks statistically on par with Harmonic (mean difference $0.17$ dB, $p=0.36$), outperforming all of the other baselines. Compared with SMN, Harmonic outperforms it by $+7.94$ dB under cosine annealing (Bonferroni-adjusted $p<{10}^{-5}$, Cohen’s $d=12.3$), winning on all six images. Additionally, the Friedman ranking across the six images confirmed Harmonic (with mean rank $=1.33$) and PM-FINER (with mean rank $=1.67$), being the top two methods under cosine annealing. Our results establish interpretable multi-harmonic and phase-modulated activations as real alternatives to the existing INR activation functions. Full article

Hereditary damping and fractional attenuation are widely used to model wave propagation in complex media, but the variational and spectral origin of the corresponding nonlocal-in-time operators is often left implicit. In this work, we derive such operators from a minimal conservative field–reservoir model. A real scalar field is coupled locally to a continuum of harmonic reservoir modes, which are then eliminated exactly. The resulting reduced dynamics is a causal wave equation with a memory-friction term acting on the field velocity. The memory kernel is generated by the reservoir coupling spectrum through a cosine-transform relation, establishing a direct spectrum-to-kernel correspondence. This relation provides both a physical interpretation of hereditary damping and a practical admissibility criterion: macroscopic attenuation and dispersion arise from the delayed back-action of unresolved internal modes, while physically admissible kernels are constrained by the non-negativity of the underlying spectral density. The framework unifies several standard damping regimes. A broadband reservoir recovers the Markovian locally damped wave equation, reservoirs with a finite characteristic time generate finite-memory relaxation and frequency-dependent dispersion, and scale-free reservoir spectra produce power-law memory kernels. In the latter case, the hereditary damping operator reduces to a Caputo-type fractional derivative, showing that fractional wave attenuation can emerge as an effective reduced dynamics rather than being postulated phenomenologically. We further analyze dispersion, attenuation, causality, stability, and admissibility conditions in terms of the reservoir spectrum. The main contribution of the work is therefore to provide a variational and spectral derivation of hereditary and fractional wave damping, linking the structure of unresolved reservoir modes to macroscopic nonlocal wave dynamics. Full article

Few-shot fault diagnosis is a challenging task in rotating machinery health monitoring because only limited labeled fault samples are available in practical industrial scenarios. Under such conditions, deep learning models are prone to overfitting and may fail to extract stable fault-sensitive features from vibration signals. Moreover, the weak fault-related components are usually coupled with operating-condition variations, background vibration, and environmental noise, which further degrades the discriminability and generalization ability of diagnostic models. To address these problems, this paper proposes a complex-valued disentangled representation learning network for few-shot fault diagnosis of rotating machinery. First, a direction-pair complex augmentation strategy is developed for triaxial vibration measurements. Two directional vibration components are selected and organized as the real and imaginary branches of a complex-valued input, which increases sample diversity under few-shot conditions. Then, a lightweight complex-valued convolution block is designed to model the coupled dynamic characteristics between different vibration directions and extract fault-sensitive representations. Furthermore, a dual-branch disentangled representation structure is developed to decompose the learned features into fault-sensitive representations and condition-related interference representations. To enhance the separability of fault embeddings under limited samples, a cosine-based disentangled representation loss is introduced, which improves intra-class compactness and inter-class discrimination while suppressing irrelevant interference information. Finally, a few-shot diagnosis strategy is constructed to identify fault categories with only a small number of labeled samples. Experimental results demonstrate that the proposed method consistently outperforms representative methods in terms of diagnostic accuracy, feature separability, and robustness, especially under extremely limited labeled samples. Full article

Phase unwrapping, the recovery of a continuous signal from measurements known only modulo $2\pi$, is a ubiquitous problem in coherent imaging, from medical MRI to radar remote sensing. In Interferometric Synthetic Aperture Radar (InSAR), phase unwrapping is both critical and computationally demanding: current methods require minutes to hours per interferogram and frequently fail on large images. We present FAUST-ADMM (Fast ADMM Unwrapping via Spectral Transforms), an algorithm that formulates phase unwrapping as a weighted $L_1$ optimization and solves it efficiently on GPU using the Alternating Direction Method of Multipliers (ADMM). Each iteration reduces to a Poisson equation solved in closed form via the Discrete Cosine Transform, followed by element-wise soft thresholding, both trivially parallel. On 500 synthetic earthquake interferograms, FAUST-ADMM achieves 99% accuracy with reference-point correction, matching SNAPHU, MCF, and PUMA, while running 10 to 100× faster. On a full three-subswath Sentinel-1 interferogram of the 2019 Ridgecrest M7.1 earthquake (∼6500 × 8500 pixels), FAUST-ADMM agrees with SNAPHU on 99.7% of pixels in 35 s, a $74\times$ speedup. Our method makes batch unwrapping of large InSAR time series practical on a single consumer GPU. Full article

Accurate and efficient oriented detection is critical for remote sensing images, yet remains challenging due to multi-scale distribution, arbitrary orientations, and stringent computational constraints of onboard platforms. To mitigate these challenges, we propose Partial-Ghost Shuffle Convolution and Gated Position-Sensitive Attention Wavelet Rotation Network (PGWave-RotNet), a lightweight wavelet-guided rotation detector that explicitly enhances multi-scale and arbitrarily oriented features while maintaining high efficiency. To reduce feature redundancy while preserving directional diversity, we design a Partial-Ghost Shuffle Convolution (PGSConv) module that integrates partial convolution with ghost shuffle. Next, to adaptively refine multi-scale and arbitrarily oriented contexts, we introduce a Gated Position-Sensitive Attention (GPSA) module with a learnable gating mechanism. To suppress aliasing and sharpen edges during upsampling, we propose a Directional-Biased Wavelet Transform Upsampling (DBWTU) module based on high-frequency wavelet reconstruction. Additionally, we develop a Weighted Cosine Angular Loss (WCAL) to improve orientation precision for square-like targets. Experiments on DOTAv1 and DIOR-R achieve 82.27% and 83.82% mAP50, outperforming existing methods. These innovations collectively enable efficient and accurate oriented detection in remote sensing. Full article

Galaxy classification is essential for understanding the formation and evolution of cosmic structures. However, faced with the explosive growth of astronomical observation data, traditional single-modality classification methods relying solely on spectroscopy or imaging have struggled to meet high-precision demands due to insufficient feature utilization and limited generalization capability. Therefore, multimodal fusion has emerged as a promising direction by leveraging information complementarity to overcome the limitations of single data sources. Accordingly, this paper proposes a model named Galaxy CosineNet (GCSNet), which integrates imaging, spectroscopic, and tabular data for high-precision galaxy classification. Specifically, the model employs dedicated encoders to process the three modalities separately and utilizes skip connections to preserve raw features. Furthermore, it incorporates a multi-head self-attention mechanism to deeply mine global cross-modal complementary information. Finally, these features are concatenated and fed into a cosine similarity classification head. Experimental results demonstrate that GCSNet achieves 97.15% accuracy in classifying star-forming, composite, active galactic nuclei (AGNs), and normal galaxies. This performance outperforms the best single-modal baseline, GaSNet, by 0.76% and mainstream multi-modal models such as MB-ISTL and the Transformer by over 1.6%. Consequently, the proposed GCSNet offers an effective and novel approach for research on automatic galaxy classification. Full article

The development of robust and efficient computational methodologies is crucial for vision-based flame detection systems. While deep learning has shown promise, many existing models are direct applications of generic architectures, lacking principled methodologies to address the inherent physical characteristics of flame, such as heat diffusion and multi-scale radiative patterns. To bridge this gap, this paper proposes FlameDet, a novel flame detection framework grounded in physics-inspired computing and multi-resolution analysis. Unlike conventional approaches, FlameDet formulates visual feature propagation through the lens of heat conduction physics. The core contribution is the Heat Diffusion Module, a computationally efficient backbone that explicitly models feature spread by solving a parameterized heat equation via discrete cosine transform. This physics-aligned design achieves a global receptive field with $O\left(N^{1.5}\right)$ complexity, processing high-resolution inputs 2× faster with 54% less memory than RT-DETR-ResNet50, while providing an interpretable computational process. Furthermore, a High–Low Frequency Analysis module is proposed, a multi-resolution computational strategy that decomposes features into low-frequency components for global context and high-frequency components for fine-grained details. To enhance contextual reasoning for small flames without computational penalty, a DSK_C3 module that employs dilated convolutions and structural re-parameterization is designed, expanding the receptive field and by 26.5%. Extensive experiments on FlameLife dataset demonstrate that FlameDet establishes a new state-of-the-art, improving the F1-score and AP50 by 3.5% and 4.0%, respectively, while maintaining superior efficiency. Full article

Current fuzzing techniques for industrial control protocols (ICPs) encounter notable challenges, including model training instability, limited sample diversity, and the inability to manage complex state dependencies in protocol interactions. To address these issues, this paper presents SD-Fuzz, a state-aware fuzzing framework that integrates a discrete denoising diffusion probabilistic model (DDPM) with an online Hidden Markov Model (HMM). The discrete DDPM is designed to generate syntactically valid and diverse protocol messages using cosine noise scheduling and Denoising Diffusion Implicit Model (DDIM) sampling, while the HMM performs unsupervised learning of state transitions from real traffic to guide the creation of logically consistent multi-step interaction sequences. The framework is evaluated on three representative Modbus/TCP slave implementations. Evaluations based on 5 h benchmark campaigns across multiple independent runs indicate that SD-Fuzz achieves a mean test case recognition rate (TCRR) of 91.3% and an HMM-inferred state transition coverage of 50.1%, exhibiting statistically significant improvements over the evaluated baselines. Furthermore, an extended 8 h vulnerability mining campaign demonstrates its capability to trigger deep-seated exceptions, including buffer overflows and protocol state violations, which are typically challenging to access using traditional stateless approaches. This work illustrates the feasibility of combining diffusion-based generation with lightweight state inference for automated vulnerability discovery in industrial control systems. Directions for future work include validation on physical programmable logic controller (PLC) hardware to acquire internal code coverage feedback. Full article

This study presents a two-phase framework for selecting clinically plausible explainable artificial intelligence (XAI) heatmaps for retinal optical coherence tomography (OCT) classification. A six-class Swin Transformer model was trained and validated using a combined dataset consisting of a subset of the public OCT-C8 dataset and private data from a Greek tertiary hospital and externally evaluated on an independent dataset from a private ophthalmological institute. Diagnostic performance was high, achieving 97% accuracy in cross-validation and 91.82% on external evaluation. In Phase 1, one ophthalmologist and one artificial intelligence (AI) specialist independently assessed 100 heatmaps per method based on visual quality and anatomical plausibility, reducing the candidate methods to three. In Phase 2, 21 specialists evaluated the selected methods across multiple cases using a five-point Likert scale reflecting agreement between highlighted regions and the model diagnosis. The proposed Token contRAST map (TRAST) achieved the highest ratings, followed by Gradient-weighted Class Activation Mapping (Grad-CAM++), while Cosine-Grad Fusion Map (CGFM) showed the lowest performance. These findings reflect clinical plausibility rather than direct model interpretability and indicate that effective XAI in OCT imaging requires not only technical performance but also structured expert evaluation. The proposed framework provides a practical approach for selecting explanation methods suitable for clinical use in ophthalmology. Full article

Adversarial examples reveal critical vulnerabilities in deep neural networks, posing significant risks in real-world deployment. In black-box settings, transferable targeted attacks rely on surrogate models but often suffer from low success rates. We argue that this limitation arises not only from surrogate-boundary overfitting but also from insufficient alignment with the target semantic space, which restricts the ability of adversarial examples to encode target-specific characteristics. To address this issue, we propose Cosine-Based Logit Alignment (CBLA), a unified framework for transferable targeted attacks. CBLA replaces the conventional cross-entropy loss with a cosine similarity objective to enhance directional alignment in logit space and alleviate gradient saturation. In addition, a semantic-invariant transformation strategy is introduced to improve structural consistency and cross-model generalization. Experiments on the ImageNet validation set demonstrate that CBLA consistently improves targeted attack success rates, achieving an average gain of 4.55% over strong baselines across multiple architectures. Full article