Residual Network Research Articles

State Surveillance and Fault Diagnosis of Distillation Columns Using Residual Network-Based Passive Acoustic Monitoring

The operational state of distillation columns significantly impacts product quality and production efficiency. However, due to the complex operation and diverse influencing factors, ensuring the safety and efficient operation of the distillation columns becomes paramount. This research combines passive acoustic monitoring with artificial intelligence techniques, proposed a technology based on residual network (ResNet), which involves the transformation of the acoustic signals emitted by three distillation columns under different operating states. The acoustic signals were initially in one-dimensional waveform format and then converted into two-dimensional Mel-Frequency Cepstral Coefficients spectrogram database using fast Fourier transform. Ultimately, this database was employed to train a ResNet for the purpose of identifying the operational states of the distillation columns. Through this approach, the operational states of distillation columns were monitored. Various faults, including flooding, entrainment, dry-tray, etc., were diagnosed with an accuracy of 98.91%. Moreover, an intermediate transitional state between normal operation and fault was identified and accurately recognized by the proposed method. Under the transitional state, the acoustic signals achieved an accuracy of 97.85% on the ResNet, which enables early warnings before faults occur, enhancing the safety of chemical production processes. The approach presents a powerful tool for the monitoring and diagnosis of chemical equipment, particularly distillation columns, ensuring the safety and efficiency.

Combining residual convolutional LSTM with attention mechanisms for spatiotemporal forest cover prediction

Understanding spatiotemporal variations in forest cover is crucial for effective forest resource management. However, existing models often lack accuracy in simultaneously capturing temporal continuity and spatial correlation. To address this challenge, we developed ResConvLSTM-Att, a novel hybrid model integrating residual neural networks, Convolutional Long Short-Term Memory (ConvLSTM) networks, and attention mechanisms. We evaluated ResConvLSTM-Att against four deep learning models: LSTM, combined convolutional neural network and LSTM (CNN-LSTM), ConvLSTM, and ResConvLSTM for spatiotemporal prediction of forest cover in Tasmania, Australia. ResConvLSTM-Att achieved outstanding prediction performance, with an average root mean square error (RMSE) of 6.9% coverage and an impressive average coefficient of determination of 0.965. Compared with LSTM, CNN-LSTM, ConvLSTM, and ResConvLSTM, ResConvLSTM-Att achieved RMSE reductions of 31.2%, 43.0%, 10.1%, and 6.5%, respectively. Additionally, we quantified the impacts of explanatory variables on forest cover dynamics. Our work demonstrated the effectiveness of ResConvLSTM-Att in spatiotemporal data modelling and prediction.

Three-dimensional PDC-SiBCN network in MA-SiBCN ceramics: Toughening-reinforcing effect and oxidation barrier

A three-dimensional (3D) precursor-derived (PDC) SiBCN network was incorporated into mechanical alloying-derived (MA) SiBCN (MA@PDC-SiBCN) ceramic through a novel hybrid organic-inorganic method, enhancing densification while providing significant reinforcement and toughening effects. The PDC-SiBCN network's submicron continuity and uniformity enable efficient load transfer between under-sintered MA-SiBCN particles, increasing strength by 130%, compared to MA-SiBCN ceramics of the same density. Additionally, as cracks penetrate the PDC-SiBCN network, crack tip deflection and bifurcation result in 57% and 352% increases in fracture toughness respectively, compared to fully sintered and under-sintered MA-SiBCN ceramics of the same density. Besides, due to synergistic oxidation characteristics, the three-dimensional PDC-SiBCN network acts as an oxidation barrier, inhibiting the heterogeneous oxidation of the encapsulated MA-SiBCN component, especially limiting the consumption of BN(C) phase at lower temperatures. The residual PDC-SiBCN network in the oxide layer serves as a rigid framework, preventing bubble growth in the molten oxide layer thereby forming a dense protective oxide layer.

DRN-DSA: A Hybrid Deep Learning Network Model for Precipitation Nowcasting using Time Series Data

Precipitation nowcasting involves short-term weather forecasting, predicting rain or snow within the next two hours. By analyzing current atmospheric conditions, it aids meteorologists in identifying weather patterns and preparing for severe events such as flooding. These nowcasts are typically displayed on geographical maps by weather services. However, the rapidly changing climate conditions make precipitation nowcasting a formidable challenge, as accurate short-term forecasts are hindered by immediate weather fluctuations. Traditional nowcasting methods, like numerical models and radar extrapolation, have limitations in delivering highly detailed and timely precipitation nowcasts. To overcome this issue, an effective solution is framed for precipitation nowcasting using a hybrid network approach named Deep Residual Network-Deep Stacked Autoencoder (DRN-DSA). Initially, the input time series data is acquired from the dataset. Thereafter, the effective technical indicators are extracted at the feature extraction stage. Later on, precipitation-type nowcasting is carried out using the proposed hybrid DRN-DSA, which is developed by incorporating a Deep Stacked Autoencoder (DSA) and Deep Residual Network (DRN). Finally, Weather nowcasting is carried out using the same proposed hybrid DSA-DRN. Moreover, when compared to other traditional models, the proposed DRN-DSA has gained superior results with a Relative Absolute Error (RAE) of 0.295, Root Mean Square Error (RMSE) of 0.154, low Mean Square Error (MSE) of 0.0236, Mean Absolute Percentage Error (MAPE) of 0.295, and False Acceptance Rate (FAR) of 0.0118.

Federated Learning With Deep Neural Networks: A Privacy-Preserving Approach to Enhanced ECG Classification.

In response to increasing data privacy regulations, this work examines the use of federated learning for deep residual networks to diagnose cardiac abnormalities from electrocardiogram (ECG) data. This approach allows medical institutions to collaborate without exchanging raw patient data. We utilize the publicly available data from the PhysioNet/Computing in Cardiology Challenge 2021, featuring diverse ECG databases, to compare the classification performance of three federated learning methods against both central training with data sharing and isolated training scenarios. We show that federated learning outperforms ECG classifiers trained in isolation. In particular, our findings demonstrate that a globally trained model fine-tuned to specific local datasets surpasses non-collaborative approaches. This shows that models trained in federation learn general features that can be tailored to specific tasks. Furthermore, federated learning almost matches the performance of central training with data sharing on out-of-distribution data from non-participating institutions. These results highlight the ability of federated learning in developing models that generalize well across diverse patient data, without the need to share data among institutions, thus addressing data privacy concerns.

High spatiotemporal resolution estimation and analysis of global surface CO concentrations using a deep learning model

Ambient carbon monoxide (CO) is a primary air pollutant that poses significant health risks and contributes to the formation of secondary atmospheric pollutants, such as ozone (O3). This study aims to elucidate global CO pollution in relation to health risks and the influence of natural events like wildfires. Utilizing artificial intelligence (AI) big data techniques, we developed a high-performance Convolutional Neural Network (CNN)-based Residual Network (ResNet) model to estimate daily global CO concentrations at a high spatial resolution of 0.07° from June 2018 to May 2021. Our model integrated the global TROPOMI Total Column of atmospheric CO (TCCO) product and reanalysis datasets, achieving desirable estimation accuracies with R-values (correlation coefficients) of 0.90 and 0.96 for daily and monthly predictions, respectively. The analysis reveals that the CO concentrations were relatively high in northern and central China, as well as northern India, particularly during winter months. Given the significant role of wildfires in increasing surface CO levels, we examined their impact in the Indochina Peninsula, the Amazon Rain Forest, and Central Africa. Our results show increases of 60.0%, 28.7%, and 40.8% in CO concentrations for these regions during wildfire seasons, respectively. Additionally, we estimated short-term mortality cases related to CO exposure in 17 countries for 2019, with China having the highest mortality cases of 23,400 (95% confidence interval: 0–99,500). Our findings highlight the critical need for ongoing monitoring of CO levels and their health implications. The daily surface CO concentration dataset is publicly available and can support future relevant sustainable studies, which is accessible at https://doi.org/10.5281/zenodo.11806178.

Multi-relation spatiotemporal graph residual network model with multi-level feature attention: A novel approach for landslide displacement prediction

Accurate prediction of landslide displacement is crucial for effective early warning of landslide disasters. While most existing prediction methods focus on time-series forecasting for individual monitoring points, there is limited research on the spatiotemporal characteristics of landslide deformation. This paper proposes a novel Multi-Relation Spatiotemporal Graph Residual Network with Multi-Level Feature Attention (MFA-MRSTGRN) that effectively improves the prediction performance of landslide displacement through spatiotemporal fusion. This model integrates internal seepage factors as data feature enhancements with external triggering factors, allowing for accurate capture of the complex spatiotemporal characteristics of landslide displacement and the construction of a multi-source heterogeneous dataset. The MFA-MRSTGRN model incorporates dynamic graph theory and four key modules: multi-level feature attention, temporal-residual decomposition, spatial multi-relational graph convolution, and spatiotemporal fusion prediction. This comprehensive approach enables the efficient analyses of multi-source heterogeneous datasets, facilitating adaptive exploration of the evolving multi-relational, multi-dimensional spatiotemporal complexities in landslides. When applying this model to predict the displacement of the Liangshuijing landslide, we demonstrate that the MFA-MRSTGRN model surpasses traditional models, such as random forest (RF), long short-term memory (LSTM), and spatial temporal graph convolutional networks (ST-GCN) models in terms of various evaluation metrics including mean absolute error (MAE = 1.27 mm), root mean square error (RMSE = 1.49 mm), mean absolute percentage error (MAPE = 0.026), and R-squared (R2 = 0.88). Furthermore, feature ablation experiments indicate that incorporating internal seepage factors improves the predictive performance of landslide displacement models. This research provides an advanced and reliable method for landslide displacement prediction.

Fault diagnosis of wind power pitch bearings based on spatiotemporal clustering and a deep attention subdomain adaptive residual network

The pitch bearing adjusts the blade pitch angle to maintain optimal windward surface exposure. However, the challenging operating conditions often result in premature localized spalling of the bearings. This study introduces an approach utilizing spatiotemporal clustering (STC) alongside a deep subdomain adaptive residual network employing channel attention to assess bearing damage. The spatiotemporal clustering feature map is created by analyzing adjacent impact times of acoustic emission signals to determine fault occurrence and severity, with the Mel-AMDSAN network employed for intelligent fault diagnosis. By integrating the “Squeeze and Excitation” (SE) module into ResNet50, domain-invariant features associated with faults are extracted, while local maximum mean discrepancy (LMMD) is utilized to align fault-type distributions across different domains. Experimental data from wind power pitch bearing test benches is utilized. The results show that the method has strong advantages and application value in terms of real-time, accuracy and cross-domain diagnosis.

A modified deep learning method for Alzheimer’s disease detection based on the facial submicroscopic features in mice

Alzheimer’s disease (AD) is a chronic disease among people aged 65 and older. As the aging population continues to grow at a rapid pace, AD has emerged as a pressing public health issue globally. Early detection of the disease is important, because increasing evidence has illustrated that early diagnosis holds the key to effective treatment of AD. In this work, we developed and refined a multi-layer cyclic Residual convolutional neural network model, specifically tailored to identify AD-related submicroscopic characteristics in the facial images of mice. Our experiments involved classifying the mice into two distinct groups: a normal control group and an AD group. Compared with the other deep learning models, the proposed model achieved a better detection performance in the dataset of the mouse experiment. The accuracy, sensitivity, specificity and precision for AD identification with our proposed model were as high as 99.78%, 100%, 99.65% and 99.44%, respectively. Moreover, the heat maps of AD correlation in the facial images of the mice were acquired with the class activation mapping algorithm. It was proven that the facial images contained AD-related submicroscopic features. Consequently, through our mouse experiments, we validated the feasibility and accuracy of utilizing a facial image-based deep learning model for AD identification. Therefore, the present study suggests the potential of using facial images for AD detection in humans through deep learning-based methods.

Dual-path aggregation transformer network for super-resolution with images occlusions and variability

While Transformer-based approaches have recently achieved notable success in super-resolution, their extensive computational requirements impede widespread practical adoption. High-resolution meteorological satellite cloud imagery is essential for weather analysis and forecasting. Enhancing image resolution through super-resolution techniques facilitates the accurate identification and localization of geographic features by meteorological systems. However, current super-resolution methods fail to restore the intricacies of cloud formations and complex regions fully. This research introduces a novel dual-path aggregation Transformer network (DPAT) tailored to enhance the super-resolution of meteorological satellite cloud images. The DPAT network adeptly captures cloud imagery's subtle details and textures, effectively addressing occlusions and the variability inherent in satellite imagery. It bolsters the model's ability to manage the complex attributes of cloud images through the introduction of the Dual-path Aggregation Self-Attention (DASA) mechanism and the Multi-scale Feature Aggregation Block (MFAB), thereby enhancing performance in processing intricate cloud features. The DASA mechanism synthesizes features across spatial, depth, and channel dimensions via a dual-path approach, thoroughly exploiting feature correlations. The MFAB, designed to supplant the multilayer perceptron, incorporates shift convolution and a multi-scale interaction block to augment feature information, compensating for the deficiency in local information absorption due to fixed receptive fields. Experimental outcomes indicate that DPAT delivers superior super-resolution outcomes. With a parameter count of only 32% of the Enhanced Deep Residual Network (EDSR) or 77% of the Image Restoration using Shift Window Transformer (SwinIR), DPAT matches SwinIR's performance on the satellite cloud dataset. Moreover, DPAT balances accuracy and parameter economy across various datasets. This technology is expected to improve image super-resolution capabilities in multiple fields such as human action recognition and industrial recognition, and indirectly improve the accuracy of image perception tasks.

Predicting an opaque bubble layer during small-incision lenticule extraction surgery based on deep learning.

This study aimed to predict the formation of OBL during femtosecond laser SMILE surgery by employing deep learning technology. This was a cross-sectional, retrospective study conducted at a university hospital. Surgical videos were randomly divided into a training (3,271 patches, 73.64%), validation (704 patches, 15.85%), and internal verification set (467 patches, 10.51%). An artificial intelligence (AI) model was developed using a SENet-based residual regression deep neural network. Model performance was assessed using the mean absolute error (E MA ), Pearson's correlation coefficient (r), and determination coefficient (R 2 ). Four distinct types of deep neural network models were established. The modified deep residual neural network prediction model with channel attention built on the PyTorch framework demonstrated the best predictive performance. The predicted OBL area values correlated well with the Photoshop-based measurements (E MA = 0.253, r = 0.831, R 2 = 0.676). The ResNet (E MA = 0.259, r = 0.798, R 2 = 0.631) and Vgg19 models (E MA = 0.31, r = 0.758, R 2 = 0.559) both displayed satisfactory predictive performance, while the U-net model (E MA = 0.605, r = 0.331, R 2 = 0.171) performed poorest. We used a panoramic corneal image obtained before the SMILE laser scan to create a unique deep residual neural network prediction model to predict OBL formation during SMILE surgery. This model demonstrated exceptional predictive power, suggesting its clinical applicability across a broad field.

Denoising diffusion probabilistic model enhanced tool condition monitoring method under imbalanced conditions

Abstract In recent years, tool condition monitoring (TCM) based on deep learning has been widely considered and achieved remarkable success. However, these methods typically require relatively large training samples to produce significant results, which are both imbalanced and rather troublesome to obtain in the practical application of TCM. To address this issue, a novel TCM method combined with multiscale permutation entropy (MPE), denoising diffusion probabilistic model (DDPM) and a residual network (ResNet) is proposed under conditions of sample imbalance. First, the one-dimensional sensing signal data is converted to a grayscale recurrence plot (RP) by minimizing the MPE of the signals in each channel. Second, combine and splice these grayscale RPs from different channels in each sample into color RPs. After that, the generated RP images using DDPM are added to the imbalanced dataset to augment the data to achieve a balanced state of the dataset. Finally, the balanced mixed data set of real and fake samples is input into a ResNet for recognition and monitoring tool conditions. TCM experiments are conducted to verify the performance of the proposed method with imbalanced dataset, and the results of experimental investigation demonstrate that the accuracy of the proposed method improved by 2%–18.8% compared to that of the other four sample augmentation methods using ResNet18 when the imbalance rate is 1:200.

Wheat Flour Discrimination Using Two-Dimensional Correlation Spectroscopy and Deep Learning.

The continuous evolution of deep learning has garnered significant attention in spectroscopy. This study focuses on identifying wheat flour, presenting a more efficient and accurate method by combining two-dimensional correlation spectroscopy (2D-COS) and deep learning techniques. A data set of 316 near-infrared (NIR) spectral samples of four types of wheat flour was collected. By applying three disparate 2D-COS techniques, i.e., synchronous, asynchronous, and integrated, we crafted 948 2D-COS images. These images, obtained by transforming the original one-dimensional spectra into 2D representations, offer richer information for deep learning analysis. The study introduced an 18-layer residual network incorporating a convolutional attention mechanism, specifically tailored for the 2D-COS analysis of wheat flour, aimed at enhancing the model's discriminative capabilities by refining the residual neural network's structure. Achieving an unprecedented recognition accuracy of 100% through methodical optimization and rigorous training on the synchronous 2D-COS data set of wheat flour, the proposed model is a testament to the efficacy of deep learning in spectroscopic analysis. To further exhibit the confluence of 2D-COS with deep learning, t-distributed stochastic neighbor embedding was employed to visualize the distinctive 2D-COS features within the deep learning architecture. Additionally, the model's performance was juxtaposed with prevailing NIR spectral recognition methods, including random forest, gradient boosting decision tree, and artificial neural network. This comparison cemented the proposed approach's superiority in wheat flour categorization. The findings of this study not only introduce a novel and efficient solution for wheat flour quality analysis but also underscore the significant potential of deep learning techniques in spectroscopy applications.

Deep learning applications for real-time and early detection of fall armyworm, African armyworm, and maize stem borer

The application of artificial intelligence for identifying Fall armyworm (Spodoptera frugiperda), African armyworm (Spodoptera exempta), and Maize stem borer (Busseola fusca) is critical due to the threats they pose to global food production. This study aims to evaluate and identify the most accurate and robust DL models in detecting and classifying these three significant agricultural pests. Seven traditional DL models: Convolutional Neural Network, Visual Geometry Group (VGG16), Residual Networks (ResNet50), MobileNetV2, InceptionV3, Deep Neural Network (DNN), and InceptionResNetV2 and the advanced You Look Only Once (YOLOv8) model were trained and tested using pest image datasets. The results showed that all traditional models except DNN had high accuracies ranging from 93.17% (InceptionResNetV2) to 99.43% (MobileNet) in training and testing, with losses ranging from 1.71% (MobileNetV2) to 24.99% (InceptionResNetV2). DNN had a slightly lower accuracy range of 55.27% to 56.39% and a loss range of 85.02% to 89.96% in training and testing. YOLOv8 emerged as the best and most robust model in the pest detection and classification tasks, achieving Precision and Recall scores ranging from 98.4% to 100% on single-class and multi-class classifications, making it highly suitable for real-world pest management applications. This research pioneers the use of DL for the classification and detection of maize stem borer, African armyworm and Fall armyworm, unique and separately addressing a critical gap in agricultural pest management in corn. With early and accurate pest identification, crop protection measures can be implemented efficiently. The findings lead to reduced crop damage and enhanced food security.

Seismic Random Noise Attenuation Using DARE U-Net

Seismic data processing plays a pivotal role in extracting valuable subsurface information for various geophysical applications. However, seismic records often suffer from inherent random noise, which obscures meaningful geological features and reduces the reliability of interpretations. In recent years, deep learning methodologies have shown promising results in performing noise attenuation tasks on seismic data. In this research, we propose modifications to the standard U-Net structure by integrating dense and residual connections, which serve as the foundation of our approach named the dense and residual (DARE U-Net) network. Dense connections enhance the receptive field and ensure that information from different scales is considered during the denoising process. Our model implements local residual connections between layers within the encoder, which allows earlier layers to directly connect with deep layers. This promotes the flow of information, allowing the network to utilize filtered and unfiltered input. The combined network mechanisms preserve the spatial information loss during the contraction process so that the decoder can locate the features more accurately by retaining the high-resolution features, enabling precise location in seismic image denoising. We evaluate this adapted architecture by applying synthetic and real data sets and calculating the peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM). The effectiveness of this method is well noted.

HOLISMOKES. XI. Evaluation of supervised neural networks for strong-lens searches in ground-based imaging surveys

While supervised neural networks have become state of the art for identifying the rare strong gravitational lenses from large imaging data sets, their selection remains significantly affected by the large number and diversity of non-lens contaminants. This work evaluates and compares systematically the performance of neural networks in order to move towards a rapid selection of galaxy-scale strong lenses with minimal human input in the era of deep, wide-scale surveys. We used multiband images from PDR2 of the Hyper-Suprime Cam (HSC) Wide survey to build test sets mimicking an actual classification experiment, with 189 securely-identified strong lenses from the literature over the HSC footprint and 70\,910 non-lens galaxies in COSMOS covering representative lens-like morphologies. Multiple networks were trained on different sets of realistic strong-lens simulations and non-lens galaxies, with various architectures and data preprocessing, mainly using the deepest $gri$-bands. Most networks reached excellent area under the Receiver Operating Characteristic (ROC) curves on the test set of 71,099 objects, and we determined the ingredients to optimize the true positive rate for a total number of false positives equal to zero or 10 $ and TPR$_ $). The overall performances strongly depend on the construction of the ground-truth training data and they typically, but not systematically, improve using our baseline residual network architecture presented in HOLISMOKES VI. TPR$_ $ tends to be higher for ResNets (simeq 10--40<!PCT!>) compared to AlexNet-like networks or G-CNNs. Improvements are found when (1) applying random shifts to the image centroids, (2) using square-root scaled images to enhance faint arcs, (3) adding $z$-band to the otherwise used $gri$-bands, or (4) using random viewpoints of the original images. In contrast, we find no improvement when adding $g - i$ difference images (where alpha is a tuned constant) to subtract emission from the central galaxy. The most significant gain is obtained with committees of networks trained on different data sets, with a moderate overlap between populations of false positives. Nearly-perfect invariance to image quality can be achieved by using realistic PSF models in our lens simulation pipeline, and by training networks either with large number of bands, or jointly with the PSF and science frames. Overall, we show the possibility to reach a $ as high as 60<!PCT!> for the test sets under consideration, which opens promising perspectives for pure selection of strong lenses without human input using the Rubin Observatory and other forthcoming ground-based surveys.

MRS‐Net: Brain tumour segmentation network based on feature fusion and attention mechanism

AbstractAccurate segmentation of brain tumor magnetic resonance imaging (MRI) is crucial for treatment planning. Addressing the challenges of complex tumor structures and inadequate cross‐channel information utilization in Unet‐based segmentation, this paper proposes the multi‐scale residual brain tumor MRI segmentation network (MRS‐Net) incorporating an attention mechanism to enhance segmentation accuracy. First, the double residual feature fusion module is utilized to enhance the fusion of feature information between different levels. Second, the Atrous Spatial Pyramid Pooling is introduced as a bridging module of the network to capture the features at different scales of the image, so as to enhance the extraction capability of the network for detailed features. Finally, the inverted residual coordinate attention module replaces the direct splicing in Unet to fuse the large feature information at each level and scale, thus enhancing the model's ability to recognize the spatial location information of brain tumors. The Dice coefficients, positive predictive values (PPVs), sensitivities (Sensitivity) and Hausdorff distance (HD), which are the four evaluation indexes, reach 84.54%, 87.43%, 88.37% and 2.248, respectively, which are improved by 1.85%, 2.11%, 2.88% and 6.0%, respectively, compared with Unet. The experimental results show that MRS‐Net achieves better brain tumor image segmentation.

Pavement crack detection based on depth-supervision FRRN model

Timely detection of early pavement distress is crucial for effective pavement maintenance. A promising approach involves the segmenting the road surface cracks area from its images. However, the previously employed full-resolution residual network (FRRN), when employing a large number of full-resolution residual units (FRRUs), is susceptible to challenges such as exploding and vanishing gradients. Addressing these issues, this paper has introduced a depth supervision mechanism into FRRN, leading to the development of the Depth Supervision FRRN (DSFRRN) model, designed specifically for accurate segmentation of cracks in road images. This model enhances nonlinear expressiveness while retaining essential features, achieving high precision in pavement crack segmentation. Through experimental validation using a dataset comprising 1565 images of damaged pavements, the result demonstrate that DSFRRN outperforms the original FRRN in terms of both F1 score and Jaccard index, accurately delineating cracks. Consequently, the automated pavement damage detection method presented herein facilitates cost-effective and efficient pavement inspection, contributing to the safeguarding of roadway surfaces and transportation infrastructure.

Mechanical equipment fault diagnosis method based on improved deep residual shrinkage network.

Fault diagnosis of mechanical equipment can effectively reduce property losses and casualties. Bearing vibration signals, as one of the effective sources of diagnostic information, are often overwhelmed by substantial environmental noise. To address this issue, we present a fault diagnosis method, CCSDRSN, which exhibits strong noise resistance. This method enhances the soft threshold function in the traditional deep residual shrinkage network, allowing it to extract useful information from the fault signal to the maximum extent, thus significantly improving diagnostic accuracy. Additionally, we have developed a novel activation function that can nonlinearly transform the time frequency map across multiple dimensions and the entire region. In pursuit of network optimization and parameter reduction, we have strategically incorporated depthwise separable convolutions, effectively replacing conventional convolutional layers. This architectural innovation streamlines the network. By verifying the effectiveness of the proposed method using Case Western Reserve University datasets, the results demonstrate that the proposed method not only possesses strong noise resistance in high noise environments but also achieves high diagnostic accuracy and good generalization performance under different load conditions.

A multi‐path residual network for image denoising based on edge prior information

AbstractEdge‐preserving denoising is important in image analysis. However, most existing methods suffer from smoothing and loss of high frequency detail at the edges. Aiming at this issue, a multi‐path residual network using edge prior information is proposed. Specifically, the edge is first recovered by a newly designed parallel edge extraction module. Then it is used as a priori to generate affine transform parameters to guide the weight distribution between noise and edge, so that the shallow feature extraction pays more attention to preserving edges. In the deep feature extraction stage, a multi‐scale attention unit is designed to provide the spatial neighbourhood information of the feature points to filter and activate the deep features, which further enhances the ability of the network to extract fine‐grained noisy features. Finally, by taking the difference between the noisy image and the noise feature obtained with residual learning, the denoised image is produced. Extensive experiments show that the PSNR is improved by 0.00–0.78 dB on grey image denoising and 0.00–2.69 dB on colour image denoising compared with others. The denoised image has a better ability to preserve image edges, high frequency details and visual perception.