Search Results (131)

Semantic segmentation of laparoscopic images requires costly pixel-level annotations, which are often unavailable for real surgical data. This gives rise to an unsupervised domain adaptation scenario, where labeled synthetic images serve as the source domain and unlabeled real images as the target. We propose a frequency-aware unsupervised domain adaptation framework to mitigate the domain gap between simulated and real laparoscopic images. Specifically, we introduce a Radial Frequency Masking module that selectively masks frequency components of real images, and employ a Mean Teacher framework to enforce consistency between high- and low-frequency representations. In addition, we propose a module called Fourier Domain Adaptation-Blend, a style transfer strategy based on low-frequency blending, and apply entropy minimization to enhance prediction confidence on the target domain. Experiments are conducted on public datasets by jointly training on simulated and real laparoscopic images. Our method consistently outperforms representative baselines. These results demonstrate the effectiveness of frequency-aware adaptation in surgical image segmentation without relying on manual annotations from the target domain. Full article

Current management strategies for sorghum aphids heavily rely on indiscriminate chemical application, leading to severe environmental consequences and impacting food safety. While precision spraying offers a viable remediation for pesticide overuse, its effectiveness depends on accurately locating and classifying pests. To address the critical challenge of segmenting small, swarming aphids in complex field environments, we propose FESW-UNet, a dual-domain attention network that integrates Fourier-enhanced attention, spatial attention, and wavelet-based downsampling into a UNet backbone. We introduce an efficient multi-scale attention (EMA) module between the encoder and decoder to enhance global context perception, enabling the model to capture more accurate relationships between global and local features in the field. In the feature extraction stage, we embed a simple attention module (SimAM) to target key infestation regions while suppressing background noise, thereby enhancing pixel-level discrimination. Furthermore, we replace conventional downsampling with Haar wavelet downsampling (HWD) to reduce resolution while preserving structural edge details. Finally, a Fourier-enhanced attention module (FEAM) is added to the skip-connection layers. By using complex-valued weights to regulate frequency-domain features, FEAM effectively fuses global low-frequency structures with local high-frequency details, thereby enhancing feature representation diversity. Experiments on the Aphid Cluster Segmentation dataset demonstrate that FESW-UNet outperforms other models, achieving an mIoU of 68.76%, mPA of 78.19%, and mF1 of 79.01%. The model also demonstrated strong adaptability on the AphidSeg-Sorghum dataset, achieving an mIoU of 81.22%, mPA of 87.97%, and mF1 of 88.60%. The proposed method offers an efficient and feasible technical solution for monitoring and controlling sorghum aphids through image segmentation, demonstrating broad application potential. Full article

This study proposes a robust chaotic encryption framework based on a Fuzzy Rule-Based Sugeno Inference (FRBSI) system, integrated with high-level security analyses. The algorithm employs a dynamic mixture of Lorenz chaotic state variables, which are numerically modeled using the Euler-Forward method to ensure computational accuracy. Unlike conventional methods, the carrier signal’s characteristics are not static; instead, its amplitude and dynamic behavior are continuously adapted through the FRBSI mechanism, driven by the instantaneous thresholds of the information signal. The security of the proposed system was rigorously evaluated through Histogram analysis, Number of Pixels Change Rate (NPCR), and Unified Average Changing Intensity (UACI) metrics, which confirmed the algorithm’s high sensitivity to plaintext variations and resistance against differential attacks. Furthermore, Key Sensitivity tests demonstrated that even a single-bit discrepancy in the receiver-side Sugeno rule base leads to a total failure in signal reconstruction, providing a formidable defense against brute-force attempts. The system’s performance was validated in the MATLAB/Simulink of R2021a version environment, where frequency and time-domain analyses were performed via oscilloscope and Fourier transforms. The results indicate that the proposed multi-layered fuzzy-chaotic structure significantly outperforms traditional encryption techniques in terms of unpredictability, structural security, and robustness. Full article

High-resolution range profile (HRRP) sequence recognition in radar automatic target recognition faces several practical challenges, including severe category imbalance, degradation of robustness under complex and variable operating conditions, and strict requirements for lightweight models suitable for real-time deployment on resource-limited platforms. To address these problems, this paper proposes a lightweight spatiotemporal fusion-based (LSTF) HRRP sequence target recognition method. First, a lightweight Transformer encoder based on group linear transformations (TGLT) is designed to effectively model temporal dynamics while significantly reducing parameter size and computation, making it suitable for edge-device applications. Second, a transform-domain spatial feature extraction network is introduced, combining the fractional Fourier transform with an enhanced squeeze-and-excitation fully convolutional network (FSCN). This design fully exploits multi-domain spatial information and enhances class separability by leveraging discriminative scattering-energy distributions at specific fractional orders. Finally, an adaptive focal loss with label smoothing (AFL-LS) is constructed to dynamically adjust class weights for improved performance on long-tail classes, while label smoothing alleviates overfitting and enhances generalization. Experiments on the MSTAR and CVDomes datasets demonstrate that the proposed method consistently outperforms existing baseline approaches across three representative scenarios. Full article

Weak-signal fluorescence channels (e.g., 4′,6-diamidino-2-phenylindole (DAPI)) often fail to provide reliable structural details due to low signal-to-noise ratio (SNR) and insufficient high-frequency information, limiting the ability of single-channel super-resolution methods to restore edge continuity and texture. This study proposes a multi-channel guided super-resolution method based on SwinIR, utilizing the high-SNR fluorescein isothiocyanate (FITC) channel as a structural reference. Dual-channel adaptation is implemented at the model input layer, enabling the window attention mechanism to fuse cross-channel correlation information and enhance the structural recovery capability of weak-signal channels. To address the loss of high-frequency information in weak-signal imaging, we introduce a frequency-domain consistency loss: this mechanism constrains spectral consistency between the predicted and true images in the Fourier domain, improving the clarity of fine-structure reconstruction. Experimental results on the DAPI channel demonstrate significant improvements: PSNR increases from 27.05 dB to 44.98 dB, and SSIM rises from 0.763 to 0.960. Visual analysis indicates that this method restores more continuous nuclear edges and weak textural details while suppressing background noise; frequency-domain results reduce the minimum resolvable feature size from approximately 1.5 μm to 0.8 μm. In summary, multi-channel structural information provides an effective and physically interpretable deep learning approach for super-resolution reconstruction of weak-signal fluorescence images. Full article

Modal decomposition, introduced by Fourier, expresses complex functions, such as sums of symmetric basis modes. However, convergence alone does not ensure structural fidelity. Discontinuities, sharp gradients, and localized features often lie outside the chosen basis’s symmetry class, producing artifacts such as the Gibbs overshoot. This study introduces a unified geometric framework for assessing when modal representations remain faithful by defining three symbolic invariants—curvature (κ), strain (τ), and compressibility (σ)—and their diagnostic ratio Γ = κ/τ. Together, these quantities measure how closely the geometry of a function aligns with the symmetry of its modal basis. The condition Γ < σ identifies the domain of structural closure: this is the region in which expansion preserves both accuracy and symmetry. Analytical demonstrations for Fourier, polynomial, and wavelet systems show that overshoot and ringing arise precisely where this inequality fails. Numerical illustrations confirm the predictive value of the invariants across discontinuous and continuous test functions. The framework reframes modal analysis as a problem of geometric compatibility rather than convergence alone, establishing quantitative criteria for closure-preserving transforms in mathematics, physics, and applied computation. It provides a general diagnostic for detecting when symmetry, curvature, and representation fall out of alignment, offering a new foundation for adaptive and structure-aware transform design. In practical terms, the invariants (κ, τ, σ) offer a diagnostic for identifying where modal systems preserve geometric structure and where they fail. Their link to symmetry arises because curvature measures structural deviation, strain measures representational effort within a given symmetry class, and compressibility quantifies efficiency. This geometric viewpoint complements classical convergence theory and clarifies why adaptive spectral methods, edge-aware transforms, multiscale PDE solvers, and learned operators benefit from locally increasing strain to restore the closure condition Γ < σ. These applications highlight the broader analytical and computational relevance of the closure framework. Full article

Parkinson’s Disease (PD) is a neurodegenerative disorder in which gait abnormalities serve as key indicators of motor impairment and disease progression. Although wearable sensor-based gait analysis has advanced, existing methods still face challenges in modeling multi-sensor spatial relationships, extracting adaptive multi-scale temporal features, and effectively integrating time–frequency information. To address these issues, this paper proposes a multi-sensor gait neural network that integrates biomechanical priors with time–frequency collaborative learning for the automatic assessment of PD gait severity. The framework consists of three core modules: (1) BGS-GAT (Biomechanics-Guided Graph Attention Network), which constructs a sensor graph based on plantar anatomy and explicitly models inter-regional force dependencies via graph attention; (2) AMS-Inception1D (Adaptive Multi-Scale Inception-1D), which employs dilated convolutions and channel attention to extract multi-scale temporal features adaptively; and (3) TF-Branch (Time–Frequency Branch), which applies Real-valued Fast Fourier Transform (RFFT) and frequency-domain convolution to capture rhythmic and high-frequency components, enabling complementary time–frequency representation. Experiments on the PhysioNet multi-channel foot pressure dataset demonstrate that the proposed model achieves 0.930 in accuracy and 0.925 in F1-score for four-class severity classification, outperforming state-of-the-art deep learning models. Full article

Early detection of breast cancer commonly relies on imaging technologies such as ultrasound, mammography and MRI. Among these, breast ultrasound is widely used by radiologists to identify and assess lesions. In this study, we developed image segmentation techniques and multiclass classification artificial intelligence (AI) tools based on pretrained models to segment lesions and detect breast cancer. The proposed workflow includes both the development of segmentation models and development of a series of classification models to classify ultrasound images as normal, benign or malignant. The pretrained models were trained and evaluated on the Breast Ultrasound Images (BUSI) dataset, a publicly available collection of grayscale breast ultrasound images with corresponding expert-annotated masks. For segmentation, images and ground-truth masks were used to pretrained encoder (ResNet18, EfficientNet-B0 and MobileNetV2)–decoder (U-Net, U-Net++ and DeepLabV3) models, including the DeepLabV3 architecture integrated with a Frequency-Domain Feature Enhancement Module (FEM). The proposed FEM improves spatial and spectral feature representations using Discrete Fourier Transform (DFT), GroupNorm, dropout regularization and adaptive fusion. For classification, each image was assigned a label (normal, benign or malignant). Optuna, an open-source software framework, was used for hyperparameter optimization and for the testing of various pretrained models to determine the best encoder–decoder segmentation architecture. Five different pretrained models (ResNet18, DenseNet121, InceptionV3, MobielNetV3 and GoogleNet) were optimized for multiclass classification. DeepLabV3 outperformed other segmentation architectures, with consistent performance across training, validation and test images, with Dice Similarity Coefficient (DSC, a metric describing the overlap between predicted and true lesion regions) values of 0.87, 0.80 and 0.83 on training, validation and test sets, respectively. ResNet18:DeepLabV3 achieved an Intersection over Union (IoU) score of 0.78 during training, while ResNet18:U-Net++ achieved the best Dice coefficient (0.83) and IoU (0.71) and area under the curve (AUC, 0.91) scores on the test (unseen) dataset when compared to other models. However, the proposed Resnet18: FrequencyAwareDeepLabV3 (FADeepLabV3) achieved a DSC of 0.85 and an IoU of 0.72 on the test dataset, demonstrating improvements over standard DeepLabV3. Notably, the frequency-domain enhancement substantially improved the AUC from 0.90 to 0.98, indicating enhanced prediction confidence and clinical reliability. For classification, ResNet18 produced an F1 score—a measure combining precision and recall—of 0.95 and an accuracy of 0.90 on the training dataset, while InceptionV3 performed best on the test dataset, with an F1 score of 0.75 and accuracy of 0.83. We demonstrate a comprehensive approach to automate the segmentation and multiclass classification of breast cancer ultrasound images into benign, malignant or normal transfer learning models on an imbalanced ultrasound image dataset. Full article

As the third generation of neural networks, Spiking Neural Networks (SNNs) simulate the event-driven processing mode of the brain, offering superior energy efficiency and biological interpretability compared to traditional deep learning. Combining the architectural strengths of Transformers with SNNs has recently demonstrated high accuracy and significant potential. SNNs process binary spikes and rich temporal information, resulting in lower computational complexity and making them particularly suitable for neuromorphic datasets. However, neuromorphic data typically involve dynamic edges and high-frequency pixel intensity changes. Capturing this frequency information is challenging for traditional spatial methods but is critical for event-driven vision. To address this, we investigate the integration of the Fast Fourier Transform (FFT) into SNNs and propose the Adaptive Spiking Fourier Neural Operator Transformer (ASFNOformer). This architecture adapts the Adaptive Fourier Neural Operator (AFNO)—originally validated in Artificial Neural Networks (ANNs)—specifically for the spiking domain. Unlike standard AFNOs, our module applies FFT across both spatial (H, W) and temporal (T) dimensions, followed by a Multi-Layer Perceptron structure (MLP) mechanism with a block-diagonal weight matrix. This design effectively captures both spatial features and temporal dynamics inherent in event streams. Furthermore, we incorporate Leaky Integrate-and-Fire (LIF) neurons optimized with Learnable Weight Parameters (LWP-LIF) to enhance temporal feature extraction and adaptivity. Experimental results on standard benchmarks indicate that our method reduces the parameter count by approximately 25%. In terms of recognition accuracy, ASFNOformer is comparable to mainstream models on static datasets and demonstrates superior performance on neuromorphic datasets by efficiently capturing frequency features. Notably, ablation studies confirm the model’s generalizability, and when using QKformer as a baseline, our method achieves state-of-the-art (SOTA) performance on the CIFAR10-DVS dataset. This work advances frequency-domain analysis in SNNs, paving the way for efficient deployment on neuromorphic hardware. Full article

Change detection (CD) in high-resolution remote sensing imagery is vital for resource monitoring and disaster assessment but faces challenges such as spatiotemporal heterogeneity, spectral variability, and computational inefficiency. This paper proposes an efficient CD method that hybridizes Convolutional Neural Networks (CNNs) and Transformers. A CNN backbone first extracts multi-level features from bi-temporal images. A Semantic Token Generator then compresses these features into compact, low-dimensional semantic tokens, reducing computational load. The core of our model is a novel cooperative encoder integrating a Spectral layer and an Attention layer. The Spectral layer enhances sensitivity to high-frequency components like edges and textures in the Fourier domain, while the Attention layer captures long-range semantic dependencies via self-attention. Furthermore, we introduce a Dynamic Tanh (DyT) module to replace conventional normalization layers, using learnable parameters to adaptively adjust activation thresholds, thereby improving training stability and computational efficiency. Comprehensive evaluations on the LEVIR-CD, WHU-CD, and DSIFN-CD benchmarks demonstrate that our method maintains high accuracy while reducing complexity, offering a practical solution for real-time CD in resource-limited environments. Full article

Change detection is a pivotal task in remote sensing information extraction, and leveraging the representation capabilities of large models has emerged as a promising direction in recent research. However, existing large-model-based change detection methods primarily focus on adaptation and fine-tuning strategies, while often overlooking the effective separation of true change information from background content. As a result, these methods still suffer from frequent false alarms and missed detections, especially in complex scenarios. To address these limitations, we propose a SAM fine-tuning adaptation change detection method based on Fourier frequency domain analysis difference reinforcement (SAM-FDN). In this method, we utilize the feature extraction capability of the SAM and adopt a low-rank fine-tuning strategy to construct the feature extraction backbone network of the model, extracting remote sensing image features at different time periods to enhance the model’s cognitive ability towards remote sensing images at different time periods. Furthermore, a Fourier Change Feature Extraction-Separation Module (FCEM) is designed based on Fourier frequency-domain analysis. This module separates high-frequency variation information from low-frequency invariant information, thereby enhancing differential features while suppressing invariant ones, which in turn contributes to more reliable and accurate remote sensing change detection (RSCD). Experiments conducted on three benchmark datasets demonstrate that SAM-FDN consistently outperforms existing state-of-the-art methods across various complex change detection scenarios. Ablation studies further confirm the effectiveness of the proposed coupling strategy between the SAM foundation model and the frequency-domain perception mechanism. In particular, the FCEM significantly improves the separation of meaningful change features and the suppression of irrelevant information, ultimately enhancing the model’s sensitivity to real changes and its overall detection performance. Full article

As optical networks continue to evolve toward higher speed and larger capacity, conventional security mechanisms relying on optoelectronic conversion are facing increasing limitations. The optical photonic firewall, as an emerging optical-layer security device, enables direct inspection in the optical domain, making its core technology—All-Optical Pattern Matching (AOPM)—a focal point of current research. This review provides a comprehensive survey of AOPM systems. It first introduces the main components of AOPM, namely symbol matching and system architectures, and analyzes their representative implementations. For low-order modulation formats such as OOK and BPSK, the review highlights matching schemes enabled by semiconductor optical amplifier (SOA) and highly nonlinear fiber (HNLF) logic gates, as well as their potential for reconfigurable extension. Building upon this foundation, the paper focuses on systems for high-order modulation formats including QPSK, 8PSK, and 16QAM, covering dimensionality-reduction-based approaches (e.g., PSA-based phase compression, squarer-based phase multiplication, constellation-mapping-based format conversion), direct symbol matching methods (e.g., phase interference, generalized XNOR, real-time Fourier transform correlation), and reconfigurable designs for multi-format adaptability. Furthermore, the review discusses optimization challenges under non-ideal conditions, such as noise accumulation, phase misalignment, and phase-locking-free operation. Finally, it outlines future directions in robust high-order modulation handling, photonic integration, and AI-driven intelligent matching, offering guidance for the development of optical-layer security technologies. Full article

Chronic low back pain (cLBP) is a pervasive and debilitating condition that can result in motor control deficits and often leads to opioid dependence. Conventional rehabilitation approaches generally rely on internally driven tasks, which fail to capture adaptive motor responses to external perturbations. This study focuses on the design and evaluation of a pneumatic-actuated active balance board integrating pneumatic artificial muscles (PAMs), electromyography (EMG), and inertial measurement units (IMUs) to assess seated postural control responses. With PAM-powered perturbations, the balance board introduces controlled challenges to evaluate postural control dynamics and motor adaptation. EMG sensors monitor muscle activity in key postural muscles, while IMU systems track movement responses. The system was evaluated through an experimental trial with 15 healthy participants performing balance tasks on both a passive and active balance board. The active balance board’s effectiveness is assessed using signal processing techniques, including root mean square (RMS) analysis, Fast Fourier Transform (FFT), autoregressive (AR) modeling, and the Welch t-test. Experimental trials were conducted with healthy participants to establish baseline performance. Results demonstrate that the active balance board successfully induces adaptive motor responses, with higher EMG activation levels compared to passive boards. Frequency-domain analyses confirm significant differences in muscle activation patterns, supporting the hypothesis that external perturbations enhance postural control retraining. The pneumatic-actuated balance board presented in this study represents a novel approach to postural control assessment that may be applied in future rehabilitation studies involving individuals with cLBP, addressing the limitations of traditional methods. Future research will focus on clinical trials with cLBP patients to further evaluate its therapeutic efficacy and long-term benefits in rehabilitation. Full article

This study adapts Fourier ptychography (FP) for high-resolution imaging in machine vision settings. We replace multi-angle illumination hardware with a single fixed light source and controlled object translation to enable a sequence of slightly shifted low-resolution frames to produce the requisite frequency-domain diversity for FP. The concept is validated in simulation using an embedded pupil function recovery algorithm to reconstruct a high-resolution complex field, recovering both amplitude and phase. For conveyor-belt transport, we introduce a lightweight preprocessing pipeline—background estimation, difference-based foreground detection, and morphological refinement—that yields robust masks and cropped inputs suitable for FP updates. The reconstructed images exhibit sharper fine structures and enhanced contrast relative to native lens imagery, indicating effective pupil synthesis without multi-LED arrays. The approach preserves compatibility with standard industrial optics and conveyor-style acquisition while reducing hardware complexity. We also discuss practical operating considerations, including blur-free capture and synchronization strategies. Full article

The accurate acquisition and real-time calculation of the attitude angle of precision firefighting extinguishing projectiles are essential for ensuring stable flight and precise extinguishing agent release. However, measuring the roll attitude angle in such projectiles is challenging due to their highly dynamic nature and environmental disturbances such as fire smoke, high temperature, and electromagnetic interference. Traditional methods for measuring attitude angles rely on multi-sensor fusion schemes, which suffer from complex structure and high cost. This paper proposes a single-gyro attitude calculation method based on micro-electromechanical inertial measurement units (MIMUs). This method integrates Fourier transform time-frequency analysis with a second-order Infinite Impulse Response (IIR) bandpass filtering algorithm optimized by dynamic coefficients. Unlike conventional fixed-coefficient filters, the proposed algorithm adaptively updates filter parameters according to instantaneous roll angular velocity, thereby maintaining tracking capability under time-varying conditions. This theoretical contribution provides a general framework for adaptive frequency-tracking filtering, beyond the specific engineering case of firefighting projectiles. Through joint time-frequency domain processing, it achieves high-precision dynamic decoupling of the roll angle, eliminating the dependency on external sensors (e.g., radar/GPS) inherent in conventional systems. This approach drastically reduces system complexity and provides key technical support for low-cost and high-reliability firefighting projectile attitude control. The research contributes to enhancing the effectiveness of urban firefighting, forest fire suppression, and public safety emergency response. Full article