Search Results (2,806)

Digital image quality is crucial for reliable analysis in applications such as medical imaging, satellite remote sensing, and video surveillance. However, traditional denoising methods struggle to balance noise removal with detail preservation and lack adaptability to various types of noise. We propose a novel three-module architecture for image denoising, comprising a generator, a dual-path-UNet-based denoiser, and a discriminator. The generator creates synthetic noise patterns to augment training data, while the dual-path-UNet denoiser uses multiple receptive field modules to preserve fine details and dense feature fusion to maintain global structural integrity. The discriminator provides adversarial feedback to enhance denoising performance. This dual-path adversarial training mechanism addresses the limitations of traditional methods by simultaneously capturing both local details and global structures. Experiments on the SIDD, DND, and PolyU datasets demonstrate superior performance. We compare our architecture with the latest state-of-the-art GAN variants through comprehensive qualitative and quantitative evaluations. These results confirm the effectiveness of noise removal with minimal loss of critical image details. The proposed architecture enhances image denoising capabilities in complex noise scenarios, providing a robust solution for applications that require high image fidelity. By enhancing adaptability to various types of noise while maintaining structural integrity, this method provides a versatile tool for image processing tasks that require preserving detail. Full article

Landslide displacement prediction is crucial for disaster mitigation, yet traditional methods often fail to capture the complex, non-stationary spatiotemporal dynamics of slope evolution. This study introduces an enhanced prediction framework that integrates multi-scale signal processing with dynamic, geology-aware graph modeling. The proposed methodology first employs the Maximum Overlap Discrete Wavelet Transform (MODWT) to denoise raw Global Navigation Satellite System (GNSS)-monitored displacement time series data, enhancing the underlying deformation features. Subsequently, a geology-aware graph is constructed, using the temporal correlation of displacement series as a practical proxy for physical relatedness between monitoring nodes. The framework’s core innovation lies in a dynamic graph optimization model with low-rank constraints, which adaptively refines the graph topology to reflect time-varying inter-sensor dependencies driven by factors like mining activities. Experiments conducted on a real-world dataset from an active open-pit mine demonstrate the framework’s superior performance. The DCRNN-proposed model achieved the highest accuracy among eight competing models, recording a Root Mean Square Error (RMSE) of 2.773 mm in the Vertical direction, a 39.1% reduction compared to its baseline. This study validates that the proposed dynamic graph optimization approach provides a robust and significantly more accurate solution for landslide prediction in complex, real-world engineering environments. Full article

To investigate the influence of different water quality conditions on the underwater transmission performance of laser communication signals, this paper systematically analyzes the absorption and scattering characteristics of the underwater laser communication channel, and constructs a transmission model of laser propagation in water, so as to explore the transmission influence mechanism under typical water quality environments. On this basis, a system of in situ measurements for underwater laser channel attenuation is designed and constructed, and several sets of experiments are carried out to verify the rationality and applicability of the model. The collected experimental data are denoised by the fusion of wavelet analysis and adaptive Kalman filtering (DWT-AKF in short) algorithm, and compared with the data measured by an underwater hyperspectral Absorption Coefficient Spectrophotometer (ACS in short), which shows that the channel attenuation coefficients of the model inversion and the measured values are in high agreement. The research results provide a reliable theoretical basis and experimental support for the performance optimization and engineering design of the underwater laser communication system. Full article

Objectives: Automatic tip localization is critical in ultrasound (US)-guided interventions. Although deep learning (DL) has been widely used for precise tip detection, existing methods are limited by the availability of real puncture data and expert annotations. Methods: To address these challenges, we propose a novel method that uses synthetic US puncture data to pre-train DL-based tip detectors, improving their generalization. Synthetic data are generated by fusing clinical US images of healthy controls with tips created using generative DL models. To ensure clinical diversity, we constructed a dataset from scans of 20 volunteers, covering 20 organs or anatomical regions, obtained with six different US machines and performed by three physicians with varying expertise levels. Tip diversity is introduced by generating a wide range of synthetic tips using a denoising probabilistic diffusion model (DDPM). This method synthesizes a large volume of diverse US puncture data, which are used to pre-train tip detectors, followed by subsequently training with real puncture data. Results: Our method outperforms MSCOCO pre-training on a clinical puncture dataset, achieving a 1.27–7.19% improvement in AP_0.1:0.5 with varying numbers of real samples. State-of-the-art detectors also show performance gains of 1.14–1.76% when applying the proposed method. Conclusions: The experimental results demonstrate that our method enhances the generalization of tip detectors without relying on expert annotations or large amounts of real data, offering significant potential for more accurate visual guidance during US-guided interventions and broader clinical applications. Full article

Single-cell RNA sequencing (scRNA-seq) has revolutionized molecular biology and genomics by enabling the profiling of individual cell types, providing insights into cellular heterogeneity. Deep learning methods have become popular in single cell analysis for tasks such as dimension reduction, cell clustering, and data imputation. In this work, we introduce DropDAE, a denoising autoencoder (DAE) model enhanced with contrastive learning, to specifically address the dropout events in scRNA-seq data, where certain genes show very low or even zero expression levels due to technical limitations. DropDAE uses the architecture of a denoising autoencoder to recover the underlying data patterns while leveraging contrastive learning to enhance group separation. Our extensive evaluations across multiple simulation settings based on synthetic data and a real-world dataset demonstrate that DropDAE not only reconstructs data effectively but also further improves clustering performance, outperforming existing methods in terms of accuracy and robustness. Full article

Magnetotelluric (MT) sounding is a crucial technique in mineral exploration. However, MT data are highly susceptible to various types of noise. Traditional data processing methods, which rely on the assumption of signal stationarity, often result in severe distortion when suppressing non-stationary noise. In this study, we propose a novel, adaptive, and less parameter-dependent signal decomposition method for MT signal denoising, based on time–frequency domain analysis and the application of modal decomposition. The method uses Variational Mode Decomposition (VMD) to adaptively decompose the MT signal into several intrinsic mode functions (IMFs), obtaining the instantaneous time–frequency energy distribution of the signal. Subsequently, robust statistical methods are introduced to extract the independent components of each IMF, thereby identifying signal and noise components within the decomposition results. Synthetic data experiments show that our method accurately separates high-amplitude non-stationary interference. Furthermore, it maintains stable decomposition results under various parameter settings, exhibiting strong robustness and low parameter dependency. When applied to field MT data, the method effectively filters out non-stationary noise, leading to significant improvements in both apparent resistivity and phase curves, indicating its practical value in mineral exploration. Full article

The rapid evolution of deep learning (DL) has fundamentally transformed the paradigm for detecting special components in agricultural products, addressing critical challenges in food safety, quality control, and precision agriculture. This comprehensive review systematically analyzes many seminal studies to evaluate cutting-edge DL applications across three core domains: contaminant surveillance (heavy metals, pesticides, and mycotoxins), nutritional component quantification (soluble solids, polyphenols, and pigments), and structural/biomarker assessment (disease symptoms, gel properties, and physiological traits). Emerging hybrid architectures—including attention-enhanced convolutional neural networks (CNNs) for lesion localization, wavelet-coupled autoencoders for spectral denoising, and multi-task learning frameworks for joint parameter prediction—demonstrate unprecedented accuracy in decoding complex agricultural matrices. Particularly noteworthy are sensor fusion strategies integrating hyperspectral imaging (HSI), Raman spectroscopy, and microwave detection with deep feature extraction, achieving industrial-grade performance ($RPD$ > 3.0) while reducing detection time by 30–100× versus conventional methods. Nevertheless, persistent barriers in the “black-box” nature of complex models, severe lack of standardized data and protocols, computational inefficiency, and poor field robustness hinder the reliable deployment and adoption of DL for detecting special components in agricultural products. This review provides an essential foundation and roadmap for future research to bridge the gap between laboratory DL models and their effective, trusted application in real-world agricultural settings. Full article

Accurate classification of breast cancer histopathology images is critical for early diagnosis and treatment planning. Yet, conventional deep learning models face significant challenges under limited annotation scenarios due to their reliance on large-scale labeled datasets. To address this, we propose Dynamic Cross-Scale Swin Transformer (DCS-ST), a robust and efficient framework tailored for histopathology image classification with scarce annotations. Specifically, DCS-ST integrates a dynamic window predictor and a cross-scale attention module to enhance multi-scale feature representation and interaction while employing a semi-supervised learning strategy based on pseudo-labeling and denoising to exploit unlabeled data effectively. This design enables the model to adaptively attend to diverse tissue structures and pathological patterns while maintaining classification stability. Extensive experiments on three public datasets—BreakHis, Mini-DDSM, and ICIAR2018—demonstrate that DCS-ST consistently outperforms existing state-of-the-art methods across various magnifications and classification tasks, achieving superior quantitative results and reliable visual classification. Furthermore, empirical evaluations validate its strong generalization capability and practical potential for real-world weakly-supervised medical image analysis. Full article

In the wake of the continuous development of the international strategic petroleum reserve system, the tonnage and quantity of oil tankers have been increasing. This trend has driven the expansion of offshore oil exploration and transportation, resulting in frequent incidents of ship oil spills. Catastrophic impacts have been exerted on the marine environment by these accidents, posing a serious threat to economic development and ecological security. Therefore, there is an urgent need for efficient and reliable methods to detect oil spills in a timely manner and minimize potential losses as much as possible. In response to this challenge, a marine radar oil film segmentation method based on feature fusion and the artificial bee colony (ABC) algorithm is proposed in this study. Initially, the raw experimental data are preprocessed to obtain denoised radar images. Subsequently, grayscale adjustment and local contrast enhancement operations are carried out on the denoised images. Next, the gray level co-occurrence matrix (GLCM) features and Tamura features are extracted from the locally contrast-enhanced images. Then, the generalized least squares (GLS) method is employed to fuse the extracted texture features, yielding a new feature fusion map. Afterwards, the optimal processing threshold is determined to obtain effective wave regions by using the bimodal graph direct method. Finally, the ABC algorithm is utilized to segment the oil films. This method can provide data support for oil spill detection in marine radar images. Full article

As the core device for protecting the safety of the pressure-bearing system, the spring full-open safety valve is prone to various forms of valve seat sealing surface damage after long-term opening and closing impact, corrosion, and medium erosion, which may lead to internal leakage. In view of the problems that the high-frequency acoustic emission signal of the internal leakage of the safety valve has, namely, a large number of energy-overlapping areas in the frequency domain, the overall signal presents broadband characteristics, large noise content, and no obvious time–frequency characteristics. A composite denoising method, IWTD, improved wavelet threshold function with dual adjustable factors, and the improved VMD algorithm is proposed. In view of the problem that the optimal values of the dual adjustment factors a and b of the function are difficult to determine manually, an improved dung beetle optimization algorithm is proposed, with the maximum Pearson coefficient as the optimization target; the optimization is performed within the value range of the dual adjustable factors a and b, so as to obtain the optimal value. In view of the problem that the key parameters K and α in VMD decomposition are difficult to determine manually, the maximum Pearson coefficient is taken as the optimization target, and the improved dung beetle algorithm is used to optimize within the value range of K and α, so as to obtain the IVMD algorithm. Based on the IVMD algorithm, the characteristic decomposition of the internal leakage acoustic emission signal occurs after the denoising of the IWTD function is performed to further improve the denoising effect. The results show that the Pearson coefficients of all types of internal leakage acoustic emission signals after IWTD-IVMD composite noise reduction are greater than 0.9, which is much higher than traditional noise reduction methods such as soft and hard threshold functions. Therefore, the IWTD-IVMD composite noise reduction method can extract more main features out of the measured spring full-open safety valve internal leakage acoustic emission signals, and has a good noise reduction effect. Feature recognition after noise reduction can provide a good evaluation for the safe operation of the safety valve. Full article

Rolling bearings serve as the most widely utilized general components in drive systems for rotating machinery, and they are susceptible to regular malfunctions. To address the performance degradation encountered by current convolutional neural network-based rolling-bearing-fault diagnosis methods due to significant noise interference and variable working conditions in industrial settings, we propose a rolling-bearing-fault diagnosis method based on dual multi-scale mechanism applicable to noisy-variable operating conditions. The suggested approach begins with the implementation of Variational Mode Decomposition (VMD) on the initial vibration signal. This is succeeded by a denoising process that utilizes the goodness-of-fit test based on the Anderson–Darling (AD) distance for enhanced accuracy. This approach targets the intrinsic mode functions (IMFs), which capture information across multiple scales, to obtain the most precise denoised signal possible. Subsequently, we introduce the Dynamic Weighted Multi-Scale Feature Convolutional Neural Network (DWMFCNN) model, which integrates two structures: multi-scale feature extraction and dynamic weighting of these features. Ultimately, the signal that has been denoised is utilized as input for the DWMFCNN model to recognize different kinds of rolling-bearing faults. Results from the experiments show that the suggested approach shows an improved denoising performance and a greater adaptability to changing working conditions. Full article

Phase-sensitive surface plasmon resonance (SPR) detection is widely employed in molecular dynamics studies and SPR imaging owing to its real-time capability, high sensitivity, and compatibility with imaging systems. A key research objective is to achieve higher measurement resolution of refractive index under optimal dynamic range conditions. We present an enhanced SPR phase imaging system combining a quad-polarization filter array for phase differential detection with a novel polarization pair, block matching, and 4D filtering (PPBM4D) algorithm to extend the dynamic range and enhance resolution. By extending the BM3D framework, PPBM4D leverages inter-polarization correlations to generate virtual measurements for each channel in the quad-polarization filter, enabling more effective noise suppression through collaborative filtering. The algorithm demonstrates 57% instrumental noise reduction and achieves 1.51 × 10⁻⁶ RIU resolution (1.333–1.393 RIU range). The system’s algorithm performance is validated through stepwise NaCl solution switching experiments (0.0025–0.08%) and protein interaction assays (0.15625–20 μg/mL). This advancement establishes a robust framework for high-resolution SPR applications across a broad dynamic range, particularly benefiting live-cell imaging and high-throughput screening. Full article

The inter-shaft bearing is an important component of aero-engine rotor systems. It works between a high-pressure rotor and a low-pressure rotor. Effective fault diagnosis of it is significant for an aero-engine. The casing vibration signals can promptly and intuitively reflect changes in the operational health status of an aero-engine’s support system. However, affected by a complex vibration transmission path and vibration of the dual-rotor, the intrinsic vibration information of the inter-shaft bearing is faced with strong noise and a dual-frequency excitation problem. This excitation is caused by the wide span of vibration source frequency distribution that results from the quite different rotational speeds of the high-pressure rotor and low-pressure rotor. Consequently, most existing fault diagnosis methods cannot effectively extract inter-shaft bearing characteristic frequency information from the casing signal. To solve this problem, this paper proposed the denoised improved envelope spectrum (DIES) method. First, an improved envelope spectrum generated by a spectrum subtraction method is proposed. This method is applied to solve the multi-source interference with wide-band distribution problem under dual-frequency excitation. Then, an improved adaptive-thresholding approach is subsequently applied to the resultant subtracted spectrum, so as to eliminate the influence of random noise in the spectrum. An experiment on a public run-to-failure bearing dataset validates that the proposed method can effectively extract an incipient bearing fault characteristic frequency (FCF) from strong background noise. Furthermore, the experiment on the inter-shaft bearing of an aero-engine test platform validates the effectiveness and superiority of the proposed DIES method. The experimental results demonstrate that this proposed method can clearly extract fault-related information from dual-frequency excitation interference. Even amid strong background noise, it precisely reveals the inter-shaft bearing’s fault-related spectral components. Full article

Point cloud denoising is essential for improving 3D data quality, yet traditional K-means methods relying on Euclidean distance struggle with non-uniform noise. This paper proposes a K-means algorithm leveraging Total Bregman Divergence (TBD) to better model geometric structures on manifolds, enhancing robustness against noise. Specifically, TBDs—Total Logarithm, Exponential, and Inverse Divergences—are defined on symmetric positive-definite matrices, each tailored to capture distinct local geometries. Theoretical analysis demonstrates the bounded sensitivity of TBD-induced means to outliers via influence functions, while anisotropy indices quantify structural variations. Numerical experiments validate the method’s superiority over Euclidean-based approaches, showing effective noise separation and improved stability. This work bridges geometric insights with practical clustering, offering a robust framework for point cloud preprocessing in vision and robotics applications. Full article

Neural networks have shown significant promise in the field of image processing, particularly for tasks such as denoising and restoration, due to their capacity to model complex nonlinear relationships between inputs and outputs. In this study, we explored the application of a complex-valued neural network—a multilayer neural network with multi-valued neurons (MLMVN)—for filtering two types of noise in digital images: additive Gaussian noise and multiplicative speckle noise. The proposed approach involves processing images as a set of overlapping patches in the frequency domain using MLMVN. Training was performed using a batch learning algorithm, which proved to be more efficient for big learning sets: it results in fewer learning epochs and a better generalization capability. Experimental results demonstrated that MLMVN achieves noise filtering quality comparable to well-established methods, such as the BM3D, Lee, and Frost filters. These findings suggest that MLMVN offers a viable framework for image denoising, particularly in scenarios where frequency domain processing is advantageous. Also, complex-valued logistic and hyperbolic tangent activation functions were used for multi-valued neurons for the first time and have shown their efficiency. Full article