Search Results (17)

The methods employed for image demosaicing and denoising play a pivotal role in image acquisition and restoration, and have been extensively studied over the past few decades. Traditionally, these tasks are performed sequentially, with demosaicing followed by denoising, or vice versa, treating each process independently. While this approach can enhance image quality, it often leads to issues such as color inaccuracies and information loss, as the outcome of the first task influences the second. Consequently, the integration of joint demosaicing and denoising (JDD) has become a focal point in recent research. Deep convolutional neural networks have shown promising results in addressing JDD challenges. This study introduces an end-to-end network, termed the Frequency-domain Features learning Network (FFNet), designed to tackle the JDD problem. Unlike conventional methods that focus on spatial domain features, FFNet utilizes frequency-domain (FD) characteristics to capture both global and local image details. Based on the vision Transformer architecture, FFNet consists of two key components: a global Fourier block (GFB), which uses global attention to determine the weights of FD parameters, and an MLP-based local Fourier block (LFB), which improves local feature extraction. These blocks are integrated with a channel attention mechanism to form the frequency-domain attention block (FAB), the core element of FFNet. Extensive experimental results on benchmark datasets demonstrate that FFNet achieves superior performance in terms of both quantitative metrics (PSNR/SSIM) and visual quality compared to existing state-of-the-art JDD methods. Furthermore, we provide a comprehensive analysis of its computational efficiency, including parameter count, FLOPs, and inference time, showing a competitive trade-off between performance and complexity. Full article

Low-dose computed tomography (LDCT) can effectively reduce ionizing radiation; however, the associated image noise and artifacts can severely compromise the accuracy of clinical diagnosis. To address the challenge of balancing noise suppression and detail preservation in LDCT images, this study proposes a deep learning (DL)-based image denoising method termed Progressive Fusion Distillation Network (PFDN). Building upon the Information Multi-distillation Network (IMDN), the proposed method incorporates a pixel attention (PA) mechanism and a progressive fusion strategy, and further designs a Pixel Parallel Extraction Block (PPEB) together with a Progressive Fusion Distillation Block (PFDB) to fully exploit multi-scale and multi-channel features, thereby optimizing the image denoising network through efficient feature separation and re-fusion. In addition, by explicitly leveraging the noise characteristics specific to LDCT images, the method establishes an end-to-end training framework suitable for medical imaging. Experimental results demonstrate that PFDN not only effectively reduces image noise and artifacts, but also enhances overall image quality while preserving diagnostically relevant image structures under the adopted evaluation setting. Full article

Electroencephalography-based motor imagery (EEG-MI) classification is a cornerstone of Brain–Computer Interface (BCI) systems, enabling the identification of motor intentions by decoding neural patterns within EEG signals. However, conventional methods, predominantly reliant on convolutional neural networks (CNNs), are proficient at extracting local temporal features but struggle to capture long-range dependencies and global contextual information. To address this limitation, we propose a Triple-path Heterogeneous Feature Collaboration Network (TPHFC-Net), which synergistically integrates three distinct temporal modeling pathways: a multi-scale Temporal Convolutional Network (TCN) to capture fine-grained local dynamics, a Transformer branch to model global dependencies via multi-head self-attention, and a Long Short-Term Memory (LSTM) network to track sequential state evolution. These heterogeneous features are subsequently fused adaptively by a dynamic gating mechanism. In addition, the model’s robustness and discriminative power are further augmented by a lightweight front-end denoising diffusion model for enhanced noisy feature representation and a back-end prototype attention mechanism to bolster the inter-class separability of non-stationary EEG features. Extensive experiments on the BCI Competition IV-2a and IV-2b datasets validate the superiority of the proposed model, achieving mean classification accuracies of 82.45% and 89.49%, respectively, on the subject-dependent MI task and significantly outperforming existing mainstream baselines. Full article

This paper proposes an efficient and automated smart healthcare communication framework that integrates a two-level filtering scheme with a multi-objective Genetic Algorithm (GA) to enhance the reliability, timeliness, and energy efficiency of Internet of Medical Things (IoMT) systems. In the first stage, physiological signals collected from heterogeneous sensors (e.g., blood pressure, glucose level, ECG, patient movement, and ambient temperature) were pre-processed using an adaptive least-mean-square (LMS) filter to suppress noise and motion artifacts, thereby improving signal quality prior to analysis. In the second stage, a GA-based optimization engine selects optimal routing paths and transmission parameters by jointly considering end-to-end delay, Signal-to-Noise Ratio (SNR), energy consumption, and packet loss ratio (PLR). The two-level filtering strategy, i.e., LMS, ensures that only denoised and high-priority records are forwarded for more processing, enabling timely delivery for supporting the downstream clinical network by optimizing the communication. The proposed mechanism is evaluated via extensive simulations involving 30–100 devices and multiple generations and is benchmarked against two existing smart healthcare schemes. The results demonstrate that the integrated GA and filtering approach significantly reduces end-to-end delay by 10%, as well as communication latency and energy consumption, while improving the packet delivery ratio by approximately 15%, as well as throughput, SNR, and overall Quality of Service (QoS) by up to 98%. These findings indicate that the proposed framework provides a scalable and intelligent communication backbone for early disease detection, continuous monitoring, and timely intervention in smart healthcare environments. Full article

Reliable operation of rolling bearings is essential for mechanical systems. Acoustic emission (AE) offers a promising approach for bearing fault detection because of its high-frequency response and strong noise-suppression capability. This study proposes an intelligent diagnostic method that combines an improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) and a convolutional neural network–bidirectional long short-term memory (CNN–BiLSTM) architecture. The method first applies wavelet denoising to AE signals, then uses ICEEMDAN decomposition followed by kurtosis-based screening to extract key fault components and construct feature vectors. Subsequently, a CNN automatically learns deep time–frequency features, and a BiLSTM captures temporal dependencies among these features, enabling end-to-end fault identification. Experiments were conducted on a bearing acoustic emission dataset comprising 15 operating conditions, five fault types, and three rotational speeds; comparative model tests were also performed. Results indicate that ICEEMDAN effectively suppresses mode mixing (average mixing rate 6.08%), and the proposed model attained an average test-set recognition accuracy of 98.00%, significantly outperforming comparative models. Moreover, the model maintained 96.67% accuracy on an independent validation set, demonstrating strong generalization and practical application potential. Full article

Hyperspectral images (HSIs) contain rich spatial–spectral information but are highly susceptible to various types of noise during the imaging process, which significantly degrades image quality and undermines the reliability of subsequent applications. To address this issue, we propose a novel end-to-end denoising framework, termed Quasi-Recursive Spectral Attention Network (QRSAN), which aims to design a feature extraction module that leverages the intrinsic characteristics of hyperspectral noise while preserving high-quality spatial and spectral information. Specifically, QRSAN introduces a Quasi-Recursive Attention Unit (QRAU) to jointly model inherent spatial–spectral dependencies, where 2D convolutions are employed for spatial feature extraction and frequency pooling is utilized for spectral representation. In addition, we develop a multi-head spectral attention mechanism to effectively capture inter-band correlations and suppress spectrally dependent noise. To further preserve fine-grained spatial structures and spectral fidelity, we design an adaptive cross-layer skip connection strategy that integrates channel-wise concatenation and transition blocks, enabling efficient feature propagation and fusion within an asymmetric encoder–decoder architecture. Extensive experiments on both synthetic and real HSI datasets demonstrate that QRSAN consistently outperforms existing methods in terms of visual quality and objective evaluation metrics, achieving superior denoising performance and generalization ability while maintaining high spatial–spectral fidelity. Full article

Ship detection in synthetic aperture radar (SAR) remote sensing imagery is of great significance in military and civilian applications. However, two factors limit detection performance: (1) a high prevalence of small-scale ship targets with limited information content and (2) interference affecting ship detection from speckle noise and land–sea clutter. To address these challenges, we propose a novel end-to-end (E2E) transformer-based SAR ship detection framework, called Flow-Aligned Nested Transformer for SAR Small Ship Detection (FANT-Det). Specifically, in the feature extraction stage, we introduce a Nested Swin Transformer Block (NSTB). The NSTB employs a two-level local self-attention mechanism to enhance fine-grained target representation, thereby enriching features of small ships. For multi-scale feature fusion, we design a Flow-Aligned Depthwise Efficient Channel Attention Network (FADEN). FADEN achieves precise alignment of features across different resolutions via semantic flow and filters background clutter through lightweight channel attention, further enhancing small-target feature quality. Moreover, we propose an Adaptive Multi-scale Contrastive Denoising (AM-CDN) training paradigm. AM-CDN constructs adaptive perturbation thresholds jointly determined by a target scale factor and a clutter factor, generating contrastive denoising samples that better match the physical characteristics of SAR ships. Finally, extensive experiments on three widely used open SAR ship datasets demonstrate that the proposed method achieves superior detection performance, outperforming current state-of-the-art (SOTA) benchmarks. Full article

Time synchronization is critical for task-oriented and time-sensitive Industrial Internet of Things (IIoT) systems. Nevertheless, achieving high-precision synchronization with low communication overhead remains a key challenge due to the constrained resources of IIoT devices. In this paper, we propose a single-timestamp time synchronization scheme that significantly reduces communication overhead by utilizing the mechanism of AP to periodically collect sensor device data. The reduced communication overhead alleviates network congestion, which is essential for achieving low end-to-end latency in synchronized IIoT networks. Furthermore, to mitigate the impact of random delay noise on clock parameter estimation, we propose a noise-robust-based Maximum Likelihood Estimation (NR-MLE) algorithm that jointly optimizes synchronization accuracy and resilience to random delays. Specifically, we decompose the collected timestamp matrix into two low-rank matrices and use gradient descent to minimize reconstruction error and regularization, approximating the true signal and removing noise. The denoised timestamp matrix is then used to jointly estimate clock skew and offset via MLE, with the corresponding Cramér–Rao Lower Bounds (CRLBs) being derived. The simulation results demonstrate that the NR-MLE algorithm achieves a higher clock parameter estimation accuracy than conventional MLE and exhibits strong robustness against increasing noise levels. Full article

The digital revolution has profoundly influenced the automotive industry, shifting the paradigm from conventional vehicles to smart cars (SCs). The SCs rely on in-vehicle communication among electronic control units (ECUs) enabled by assorted protocols. The Controller Area Network (CAN) serves as the de facto standard for interconnecting these units, enabling critical functionalities. However, inherited non-delineation in SCs— transmits messages without explicit destination addressing—poses significant security risks, necessitating the evolution of an astute and resilient self-defense mechanism (SDM) to neutralize cyber threats. To this end, this study introduces a lightweight intrusion mitigation mechanism based on an adaptive momentum-based deep denoising autoencoder (AM-DDAE). Employing real-time CAN bus data from renowned smart vehicles, the proposed framework effectively reconstructs original data compromised by adversarial activities. Simulation results illustrate the efficacy of the AM-DDAE-based SDM, achieving a reconstruction error (RE) of less than 1% and an average execution time of 0.145532 s for data recovery. When validated on a new unseen attack, and on an Adversarial Machine Learning attack, the proposed model demonstrated equally strong performance with RE < 1%. Furthermore, the model’s decision-making capabilities were analysed using Explainable AI techinques such as SHAP and LIME. Additionally, the scheme offers applicable deployment flexibility: it can either be (a) embedded directly into individual ECU firmware or (b) implemented as a centralized hardware component interfacing between the CAN bus and ECUs, preloaded with the proposed mitigation algorithm. Full article

Seismic signal processing often relies on general convolutional neural network (CNN)-based models, which typically focus on features in the time domain while neglecting frequency characteristics. Moreover, down-sampling operations in these models tend to cause the loss of critical high-frequency details. To this end, we propose a feedback information distillation network (FID-N) in the non-subsampled contourlet transform (NSCT) domain to remarkably suppress seismic noise. The method aims to enhance denoising performance by preserving the fine-grained details and frequency characteristics of seismic data. The FID-N mainly consists of a two-path information distillation block used in a recurrent manner to form a feedback mechanism, carrying an output to correct previous states, which fully exploits competitive features from seismic signals and effectively realizes the signal restoration step by step across time. Additionally, the NSCT has an excellent high-frequency response and powerful curve and surface description capabilities. We suggest converting the noise suppression problem into NSCT coefficient prediction, which maintains more detailed high-frequency information and promotes the FID-N to further suppress noise. Extensive experiments on both synthetic and real seismic datasets demonstrated that our method significantly outperformed the SOTA methods, particularly in scenarios with low signal-to-noise ratios and in recovering high-frequency components. Full article

Liver cancer is a major global health challenge, significantly contributing to mortality rates. The accurate segmentation of liver and tumors from clinical CT images plays a crucial role in selecting therapeutic strategies for liver disease and treatment monitoring but remains challenging due to liver shape variability, proximity to other organs, low contrast between tumors and healthy tissues, and unclear lesion boundaries. To address these challenges, we propose the Deep Residual Dual-Attention Network (DRDA-Net), a novel model for end-to-end liver and tumor segmentation. DRDA-Net integrates a Residual UNet architecture with dual-attention mechanisms, multi-scale tile and patch extraction, and an Ensemble method. The dual-attention mechanisms enhance focus on key regions, addressing variations in size, shape, and intensity, while the multi-scale approach captures fine details and broader contexts. Additionally, we introduce a unique pre-processing pipeline employing a two-channel denoising technique using convolutional neural networks (CNNs) and stationary wavelet transforms (SWTs) to reduce noise while preserving structural details. Evaluated on the 3DIRCADb dataset, DRDA-Net achieved Dice scores of 97.03% and 75.4% for liver and tumor segmentation, respectively, outperforming state-of-the-art methods. These results demonstrate the model’s effectiveness in overcoming segmentation challenges and highlight its potential to improve liver cancer diagnostics and treatment planning. Full article

Transformer-based methodologies in object detection have recently piqued considerable interest and have produced impressive results. DETR, an end-to-end object detection framework, ingeniously integrates the Transformer architecture, traditionally used in NLP, into computer vision for sequence-to-sequence prediction. Its enhanced variant, DINO, featuring improved denoising anchor boxes, has showcased remarkable performance on the COCO val2017 dataset. However, it often encounters challenges when applied to scenarios involving small object detection. Thus, we propose an innovative method for feature enhancement tailored to recursive prediction tasks, with a particular emphasis on augmenting small object detection performance. It primarily involves three enhancements: refining the backbone to favor feature maps that are more sensitive to small targets, incrementally augmenting the number of queries for small objects, and advancing the loss function for better performance. Specifically, The study incorporated the Switchable Atrous Convolution (SAC) mechanism, which features adaptable dilated convolutions, to increment the receptive field and thus elevate the innate feature extraction capabilities of the primary network concerning diminutive objects. Subsequently, a Recursive Small Object Prediction (RSP) module was designed to enhance the feature extraction of the prediction head for more precise network operations. Finally, the loss function was augmented with the Normalized Wasserstein Distance (NWD) metric, tailoring the loss function to suit small object detection better. The efficacy of the proposed model is empirically confirmed via testing on the VISDRONE2019 dataset. The comprehensive array of experiments indicates that our proposed model outperforms the extant DINO model in terms of average precision (AP) small object detection. Full article

Synthetic aperture radar (SAR) images are inevitably interspersed with speckle noise due to their coherent imaging mechanism, which greatly hinders subsequent related research and application. In recent studies, deep learning has become an effective tool for despeckling remote sensing images. However, preserving more texture details while removing speckle noise remains a challenging task in the field of SAR image despeckling. Furthermore, most despeckling algorithms are designed specifically for a specific look and seriously lack generalizability. Therefore, in order to remove speckle noise in SAR images, a novel end-to-end frequency domain decomposition network (SAR−FDD) is proposed. The method first performs frequency domain decomposition to generate high-frequency and low-frequency information. In the high-frequency branch, a mean filter is employed to effectively remove noise. Then, an interactive dual-branch framework is utilized to learn the details and structural information of SAR images, effectively reducing speckles by fully utilizing features from different frequencies. In addition, a blind denoising model is trained to handle noisy SAR images with unknown noise levels. The experimental results demonstrate that the SAR−FDD achieves good visual effects and high objective evaluation metrics on both simulated and real SAR test sets (peak signal-to-noise ratio (PSNR): 27.59 ± 1.57 and structural similarity index (SSIM): 0.78 ± 0.05 for different speckle noise levels), demonstrating its strong denoising performance and ability to preserve edge textures. Full article

Deep convolution neural networks (CNNs) have become popular for image denoising due to their robust learning capabilities. However, many methods tend to increase the receptive field to improve performance, which leads to over-smoothed results and loss of critical high-frequency information such as edges and texture. In this research, we introduce an innovative end-to-end denoising network named the noise gating dynamic convolutional network (NGDCNet). By integrating dynamic convolution and noise gating mechanisms, our approach effectively reduces noise while retaining finer image details. Through a series of experiments, we conduct a comprehensive evaluation of NGDCNet by comparing it quantitatively and visually against state-of-the-art denoising methods. Additionally, we provide an ablation study to analyze the contributions of dynamic convolutional blocks and noise gating blocks. Our experimental findings demonstrate that NGDCNet excels in noise reduction while preserving essential texture information. Full article

The distribution of real noise in images can disrupt the inherent symmetry present in many natural visuals, thus making its effective removal a paramount challenge. However, traditional denoising methods often require tedious manual parameter tuning, and a significant portion of deep learning-driven techniques have proven inadequate for real noise. Moreover, the efficiency of end-to-end algorithms in restoring symmetrical patterns in noisy images remains questionable. To harness the principles of symmetry for improved denoising, we introduce a dual deep learning model with a focus on preserving and leveraging symmetrical patterns in real images. Our methodology operates in two stages. In the first, we estimate the noise level using a four-layer neural network, thereby aiming to capture the underlying symmetrical structures of the original image. To enhance the extraction of symmetrical features and overall network performance, a dual attention mechanism is employed before the final convolutional layer. This innovative module adaptively assigns weights to features across different channels, thus emphasizing symmetry-preserving elements. The subsequent phase is devoted to non-blind denoising. It integrates the estimated noise level and the original image, thus targeting the challenge of denoising while preserving symmetrical patterns. Here, a multi-scale architecture is used, thereby amalgamating image features into two branches. The first branch taps into dilation convolution, thus amplifying the receptive field without introducing new parameters and making it particularly adept at capturing broad symmetrical structures. In contrast, the second branch employs a standard convolutional layer to focus on finer symmetrical details. By harnessing varied receptive fields, our method can recognize and restore image symmetries across different scales. Crucial skip connections are embedded within this multi-scale setup, thus ensuring that symmetrical image data is retained as the network deepens. Experimental evaluations, conducted on four benchmark training sets and 12 test datasets, juxtaposed with over 20 contemporary models based on the peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) metrics, underscore our model’s prowess in not only denoising but also in preserving and accentuating symmetrical elements, thereby setting a new gold standard in the field. Full article