Search Results (551)

This paper proposes a novel image tone-mapping framework that incorporates meta-learning, a psychophysical model, Bayesian optimization, and light-field virtual diffraction. First, we formalize the virtual diffraction process as a mathematical operator defined in the frequency domain to reconstruct high-dynamic-range (HDR) images through phase modulation, enabling the precise control of image details and contrast. In parallel, we apply the Stevens power law to simulate the nonlinear luminance perception of the human visual system, thereby adjusting the overall brightness distribution of the HDR image and improving the visual experience. Unlike existing methods that primarily emphasize structural fidelity, the proposed method strikes a balance between perceptual fidelity and visual naturalness. Secondly, an adaptive parameter tuning system based on Bayesian optimization is developed to conduct optimization of the Tone Mapping Quality Index (TMQI), quantifying uncertainty using probabilistic models to approximate the global optimum with fewer evaluations. Furthermore, we propose a task-distribution-oriented meta-learning framework: a meta-feature space based on image statistics is constructed, and task clustering is combined with a gated meta-learner to rapidly predict initial parameters. This approach significantly enhances the robustness of the algorithm in generalizing to diverse HDR content and effectively mitigates the cold-start problem in the early stage of Bayesian optimization, thereby accelerating the convergence of the overall optimization process. Experimental results demonstrate that the proposed method substantially outperforms state-of-the-art tone-mapping algorithms across multiple benchmark datasets, with an average improvement of up to 27% in naturalness. Furthermore, the meta-learning-guided Bayesian optimization achieves two- to five-fold faster convergence. In the trade-off between computational time and performance, the proposed method consistently dominates the Pareto frontier, achieving high-quality results and efficient convergence with a low computational cost. Full article

This paper proposes a bearing fault diagnosis method based on the Dempster–Shafer evidence fusion of cloud model memberships from multi-channel data, which provides an explicable calculation process and a final result. Firstly, vibration signals from the drive end and fan end of the rolling bearing are used as dual-channel data sources to extract multi-dimensional features from time and frequency domains. Then, cloud models are employed to build models for each feature under different conditions, utilizing three digital characteristic parameters to characterize the distribution and uncertainty of features under different operating conditions. Thus, the membership degree vectors of test samples from two channels can be calculated using reference models. Subsequently, D-S evidence theory is applied to fuse membership degree vectors of the two channels, effectively enhancing the robustness and accuracy of the diagnosis. Experiments are conducted on the rolling bearing fault dataset from Case Western Reserve University. Results demonstrate that the proposed method achieves an accuracy of 96.32% using evidence fusion of the drive-end and fan-end data, which is obviously higher than that seen in preliminary single-channel diagnosis. Meanwhile, the final results can give suggestions of the possibilities of anther, which is benefit for technicists seeking to investigate the actual situation. Full article

In recent years, drone-based wildfire detection technology has advanced rapidly, yet existing methods still encounter numerous challenges. For instance, high background complexity leads to frequent false positives and false negatives in models, which struggle to accurately identify both small-scale fire points and large-scale wildfires simultaneously. Furthermore, the complex model architecture and substantial parameter count hinder lightweight deployment requirements for drone platforms. To this end, this paper presents a lightweight drone-based wildfire detection model, DFE-YOLO. This model utilizes dynamic frequency domain enhancement technology to resolve the aforementioned challenges. Specifically, this study enhances small object detection capabilities through a four-tier detection mechanism; improves feature representation and robustness against interference by incorporating a Dynamic Frequency Domain Enhancement Module (DFDEM) and a Target Feature Enhancement Module (C2f_CBAM); and significantly reduces parameter count via a multi-scale sparse sampling module (MS3) to address resource constraints on drones. Experimental results demonstrate that DFE-YOLO achieves mAP50 scores of 88.4% and 88.0% on the Multiple lighting levels and Multiple wildfire objects Synthetic Forest Wildfire Dataset (M4SFWD) and Fire-detection datasets, respectively, whilst reducing parameters by 23.1%. Concurrently, mAP50-95 reaches 50.6% and 63.7%. Comprehensive results demonstrate that DFE-YOLO surpasses existing mainstream detection models in both accuracy and efficiency, providing a reliable solution for wildfire monitoring via unmanned aerial vehicles. Full article

In image splicing tamper detection, forgery operations simultaneously introduce macroscopic semantic inconsistencies and microscopic tampering artifacts. Conventional methods often treat semantic understanding and low-level artifact perception as separate tasks, which impedes their effective synergy. Meanwhile, frequency-domain information, a crucial clue for identifying traces of tampering, is frequently overlooked. However, a simplistic fusion of frequency-domain and spatial features can lead to feature conflicts and information redundancy. To resolve these challenges, this paper proposes an Adaptive Multi-Scale Edge-Aware Network (AMSEANet). This network employs a synergistic enhancement cascade architecture, recasting semantic understanding and artifact perception as a single, frequency-aware process guided by deep semantics. It leverages data-driven adaptive filters to precisely isolate and focus on edge artifacts that signify tampering. Concurrently, the dense fusion and enhancement of cross-scale features effectively preserve minute tampering clues and boundary details. Extensive experiments demonstrate that our proposed method achieves superior performance on several public datasets and exhibits excellent robustness against common attacks, such as noise and JPEG compression. Full article

Data augmentation is crucial for electric motor fault diagnosis and lifetime prediction. However, the diversity of operating conditions and the challenge of augmenting small datasets often limit existing models. To address this, we propose an enhanced TimeGAN framework that couples the original architecture with transformer modules to jointly exploit time- and frequency-domain information to improve the fidelity of synthetic motor signals. The method fuses raw waveforms, envelope features, and instantaneous phase-change cues to strengthen temporal representation learning. The generator further incorporates frequency-domain descriptors and adaptively balances time–frequency contributions through learnable weighting, thereby improving generative performance. In addition, a state-conditioning mechanism (via explicit state annotations) enables controlled synthesis across distinct operating states. Comprehensive evaluations—including PCA and t-SNE visualizations, distance metrics such as DTW and FID, and downstream classifier tests—demonstrate strong adaptability and robustness on both public and in-house datasets, substantially enhancing the quality of generated time series. Full article

A smart device unlocking scheme based on ultrasonic gesture recognition is proposed, allowing users to unlock their devices by customizing the unlock code through gesture movements. This method utilizes ultrasound to detect multiple consecutive gestures, identifying micro-features within these gestures for authentication. To enhance recognition accuracy, an unsupervised segmentation algorithm is employed to accurately segment the gesture feature region and extract the time-frequency domain data of the gestures. Additionally, two-stage data enhancement techniques are applied to generate user-specific data based on a small sample size. Finally, the user-specific model is deployed to mobile devices via transfer learning for on-device, real-time inference. Experimental validation on a commercial smartphone (Redmi K50) demonstrates that the entire authentication pipeline, from signal acquisition to decision, processes 8 types of gestures in a sequence in sequence in approximately 1.2 s, with the core model inference taking less than 50 milliseconds. This ensures that the raw biometric data (ultrasonic echoes) and the recognition results never leave the user’s device during authentication, thereby safeguarding privacy. It is important to note that while model training is performed offline on a server to leverage greater computational resources for personalization, the deployed system operates fully in real time on the edge device. Experimental results demonstrate that our system achieves accurate and robust identity verification, with an average five-fold cross-validation accuracy rate of up to $93.56\%$, and it shows robustness across different environments. Full article

With the development of artificial intelligence neural networks, semantic segmentation has received more and more attention in the field of ocean engineering, especially in the fields of unmanned vessels and drones. However, challenges such as limited open ocean datasets, insufficient feature extraction for segmentation networks, pixel pairing problem, and frequency-domain obfuscation still exist. To address these issues, we propose USVS-Net, a high-performance segmentation network for segmenting USV feasible domains and surface obstacles. To overcome the pixel pairing confusion problem, a Global Channel-Spatial Attention module (GCSA) is designed in this paper, which enhances feature interactions, suppresses redundant features, and improves pixel matching accuracy through channel shuffling strategy and large kernel spatial attention. In addition, a median-enhanced channel-spatial attention (MECS) module is proposed to enhance edge details and suppress noise by fusing the median, mean, and maximum values to facilitate cross-scale feature interactions. For evaluation, a dataset USV-DATA containing images of marine obstacles is constructed. Experiments show that USVS-Net outperforms SOTA with mIoU reaching 81.71% and mPA reaching 90.18%, which is a significant improvement over the previous methods. These findings indicate that USVS-Net has high accuracy and robustness and can provide valuable support for autonomous navigation of unmanned vessels. Full article

Pipeline leak detection through acoustic signal measurement faces critical challenges, including insufficient utilization of time-frequency domain features, poor adaptability to noisy environments, and inadequate exploitation of frequency-domain prior knowledge in existing deep learning approaches. This paper proposes a Multi-Scale Frequency-Aware Transformer (MSFAT) architecture that integrates measurement-based acoustic signal analysis with artificial intelligence techniques. The MSFAT framework consists of four core components: a frequency-aware embedding layer that achieves joint representation learning of time-frequency dual-domain features through parallel temporal convolution and frequency transformation, a multi-head frequency attention mechanism that dynamically adjusts attention weights based on spectral distribution using frequency features as modulation signals, an adaptive noise filtering module that integrates noise detection, signal enhancement, and adaptive fusion functions through end-to-end joint optimization, and a multi-scale feature aggregation mechanism that extracts discriminative global representations through complementary pooling strategies. The proposed method addresses the fundamental limitations of traditional measurement-based detection systems by incorporating domain-specific prior knowledge into neural network architecture design. Experimental validation demonstrates that MSFAT achieves 97.2% accuracy and an F1-score, representing improvements of 10.5% and 10.9%, respectively, compared to standard Transformer approaches. The model maintains robust detection performance across signal-to-noise ratio conditions ranging from 5 to 30 dB, demonstrating superior adaptability to complex industrial measurement environments. Ablation studies confirm the effectiveness of each innovative module, with frequency-aware mechanisms contributing most significantly to the enhanced measurement precision and reliability in pipeline leak detection applications. Full article

Transformer reliability is crucial to grid security, with core loosening a common fault. This paper proposes a transformer core loosening fault diagnosis method based on a fusion feature extraction approach and Categorical Boosting (CatBoost) optimized by the Crested Porcupine Optimizer (CPO) algorithm. Firstly, the audio signal is decomposed into six Intrinsic Mode Functions (IMF) components through Variational Mode Decomposition (VMD). This paper utilizes Gaussian membership functions to quantify the energy proportion, central frequency, and kurtosis of IMF and constructs a fuzzy entropy discrimination function. Then, the IMF noise components are removed through an adaptive threshold. Subsequently, the denoised signal undergoes a wavelet packet transform instead of a short-time Fourier transform to optimize Mel-frequency cepstral coefficients (WPT-MFCC), combining time-domain statistical features and frequency-band energy distribution to form a 24-dimensional fusion feature. Finally, the CatBoost algorithm is employed to validate the effects of different feature schemes. The CPO is introduced to optimize its iteration number, learning rate, tree depth, and random strength parameters, thereby enhancing overall performance. The CPO-optimized CatBoost model had 99.0196% fault recognition accuracy in experimental testing, 15% better than the standard CatBoost. Accuracy exceeded 90% even under extreme 0 dB noise. This method makes fault diagnosis more accurate and reliable. Full article

In recent years, Mamba has gained increasing importance in the field of image restoration, gradually outperforming traditional convolutional neural networks (CNNs) and Transformers. However, the existing Mamba-based networks mainly focus on capturing global contextual relationships and neglect the crucial impact of local feature interactions on restoration performance in low-light environments. These environments inherently require the joint optimization of multi-scale spatial dependencies and frequency-domain characteristics. The traditional CNNs and Transformers face challenges in modeling long-range dependencies, while State Space Models (SSMs) in Mamba demonstrate proficiency in sequential modeling yet exhibit limitations in fine-grained feature extraction. To address the limitations of existing methods in capturing global degradation patterns, this paper proposes a novel global-to-local feature extraction framework through systematic Mamba integration. The Low-Frequency Mamba Block (LFMBlock) is introduced first to perform refined feature extraction in the low-frequency domain. The High-Frequency Guided Enhancement Block (HFGBlock) is used, which utilizes low-frequency priors to compensate for texture distortions in high-frequency components. Comprehensive experiments on multiple benchmark datasets show that the Global-to-Local Mamba architecture achieves superior performance in low-light restoration and image enhancement. It significantly outperforms state-of-the-art methods in both quantitative metrics and visual quality preservation, especially in recovering edge details and suppressing noise amplification under extreme illumination conditions. The hierarchical design effectively bridges global structural recovery with local texture refinement, setting a new paradigm for frequency-aware image restoration. Full article

In open-network environments of smart buildings and urban infrastructure, abnormal traffic from security and energy monitoring systems is critical for operational safety and decision reliability. We can develop malware that exploits building automation protocols to simulate attacks involving the falsification or modification of chiller controller commands, thereby endangering the entire network infrastructure. Intrusion detection systems rely on abundant labeled abnormal traffic data to detect attack patterns, improving network system reliability. However, transmitting such data faces two major challenges: single-feature representations fail to capture comprehensive traffic features, limiting the information representation for artificial intelligence (AI)-based detection models, and unconcealed abnormal traffic is easily intercepted by firewalls or intrusion detection systems, hindering cross-departmental sharing. Existing methods struggle to balance feature integrity and transmission stealth, often sacrificing one for the other or relying on easily detectable spatial-domain steganography. To address these gaps, we propose a multi-channel imaging hiding method that reconstructs abnormal traffic into multi-channel images by combining three mappings to generate grayscale images that depict traffic state transitions, dynamic trends, and internal similarity, respectively. These images are combined to enhance feature representation and embedded into frequency-domain adversarial examples, enabling evasion of security devices while preserving traffic integrity. Experimental results demonstrate that our method captures richer information than single-representation approaches, achieving a PSNR of 44.5 dB (a 6.0 dB improvement over existing methods) and an SSIM of 0.97. The high-fidelity reconstructions enabled by these gains facilitate the secure and efficient sharing of abnormal traffic data, thereby enhancing AI-driven security in smart buildings. Full article

Objective evaluation of pilot fatigue is crucial for enhancing aviation safety. Although electroencephalography (EEG) is regarded as an effective tool for recognizing pilot fatigue, the direct application of deep learning models to raw EEG signals faces significant challenges due to issues such as massive data volume, excessively long training time, and model overfitting. Moreover, existing feature-based methods often suffer from data redundancy due to the lack of effective feature and channel selections, which compromises the model’s recognition efficiency and accuracy. To address these issues, this paper proposes a framework, named ASFT-Transformer, for fast and accurate detection of pilot fatigue. This framework first extracts time-domain and frequency-domain features from the four EEG frequency bands. Subsequently, it introduces a feature and channel selection strategy based on one-way analysis of variance and support vector machine (ANOVA-SVM) to identify the most fatigue-relevant features and pivotal EEG channels. Finally, the FT-Transformer (Feature Tokenizer + Transformer) model is employed for classification based on the selected features, transforming the fatigue recognition problem into a tabular data classification task. EEG data is collected from 32 pilots before and after actual simulator training to validate the proposed method. The results show that ASFT-Transformer achieved average accuracies of 97.24% and 87.72% based on cross-clip data partitioning and cross-subject data partitioning, which were significantly superior to several mainstream machine learning and deep learning models. Under the two types of cross-validation, the proposed feature and channel selection strategy not only improved the average accuracy by 2.45% and 8.07%, respectively, but also drastically reduced the average training time from above 1 h to under 10 min. This study offers civil aviation authorities and airline operators a tool to manage pilot fatigue objectively and effectively, thereby contributing to flight safety. Full article

The high spatiotemporal resolution of optically pumped magnetometers (OPMs) makes them an essential tool for functional brain imaging, enabling accurate recordings of neuronal activity. However, physiological signals such as eye blinks and cardiac activity overlap with neural magnetic signals in the frequency domain, resulting in contamination and creating challenges for the observation of brain activity and the study of neurological disorders. To address this problem, an automatic physiological artifact removal method based on OPM magnetic reference signals and a channel attention mechanism is proposed. The randomized dependence coefficient (RDC) is employed to evaluate the correlation between independent components and reference signals, enabling reliable recognition of artifact components and the construction of training and testing datasets. A channel attention mechanism is subsequently introduced, which fuses features from global average pooling (GAP) and global max pooling (GMP) layers through convolution to establish a data-driven automatic recognition model. The backbone network is further optimized to enhance performance. Experimental results demonstrate a strong correlation between the magnetic reference signals and artifact components, confirming the reliability of magnetic signals as artifact references for OPM-MEG. The proposed model achieves an artifact recognition accuracy of 98.52% and a macro-average score of 98.15%. After artifact removal, both the event-related field (ERF) responses and the signal-to-noise ratio (SNR) are significantly improved. Leveraging the flexible and modular characteristics of OPM-MEG, this study introduces an artifact recognition framework that integrates magnetic reference signals with an attention mechanism. This approach enables highly accurate automatic recognition and removal of OPM-MEG artifacts, paving the way for real-time, automated data analysis in both scientific research and clinical applications. Full article

To address the technical challenges in the educational domain named entity recognition (NER), such as ambiguous entity boundaries and difficulties with nested entity identification, this study proposes an enhanced semantic BERT model (ES-BERT). The model innovatively adopts an education domain, vocabulary-assisted semantic enhancement strategy that (1) applies the term frequency–inverse document frequency (TF-IDF) algorithm to weight domain-specific terms, and (2) fuses the weighted lexical information with character-level features, enabling BERT to generate enriched, domain-aware, character–word hybrid representations. A complete bidirectional long short-term memory-conditional random field (BiLSTM-CRF) recognition framework was established, and a novel focal loss-based joint training method was introduced to optimize the process. The experimental design employed a three-phase validation protocol, as follows: (1) In a comparative evaluation using 5-fold cross-validation on our proprietary computer-education dataset, the proposed ES-BERT model yielded a precision of 90.38%, which is higher than that of the baseline models; (2) Ablation studies confirmed the contribution of domain-vocabulary enhancement to performance improvement; (3) Cross-domain experiments on the 2016 knowledge base question answering datasets and resume benchmark datasets demonstrated outstanding precision of 98.41% and 96.75%, respectively, verifying the model’s transfer-learning capability. These comprehensive experimental results substantiate that ES-BERT not only effectively resolves domain-specific NER challenges in education but also exhibits remarkable cross-domain adaptability. Full article

Wind turbines serve as core equipment for renewable energy development and utilization. Tower high-strength bolts are critical load-bearing components of wind turbines, making precise detection and early identification of their failures of significant importance. By addressing the challenges of extracting failure features from tower high-strength bolts and the insufficient generalization capability of traditional intelligent diagnostic models, this study achieves the precise detection and early identification of bolt-loosening failures. This study independently constructed a simulation platform to collect bolt vibration data and proposed a diagnostic model—BKA-CNN-GRU—based on a CNN-GRU architecture enhanced by the Black-winged Kite Algorithm (BKA). This approach enables precise detection and early identification of high-strength bolt-loosening failures. The specific research approach involved first establishing experimental conditions with varying bolt tightening levels to capture time-domain vibration signals. Subsequently, Fast Fourier Transform (FFT) was applied to convert these signals into frequency-domain data. Building upon this, BKA optimization was employed to refine key hyperparameters of the CNN-GRU model, thereby enhancing diagnostic performance. Experimental results demonstrate that, under the same dataset, the BKA-CNN-GRU model achieves an average accuracy of 99.31%, significantly outperforming comparison models such as CNN-GRU. Furthermore, the model exhibits robust performance and strong generalization capabilities under varying sample proportions and small-sample conditions. Full article