Noise Reduction Model Research Articles

Effective signal denoising with adaptive polynomials

Abstract The study presents an informatic model for signal processing and noise reduction, harnessing the assistance of mathematics. Advanced noise reduction algorithms are applied to improve signal quality. A MATLAB program was developed integrating algorithms designed to reduce noise regardless of its parameter. Employment of adaptive polynomials was also utilized to enhance the reduction of noise interference. All algorithms within the model have been thoroughly automated, ensuring seamless execution. The results demonstrate significant improvements in signal clarity, with filtered signals closely matching the expected noiseless signals. One of the model’s aims was to achieve a significant reduction in noise emitted by rapidly moving objects, an aim that was effectively achieved. The model is applicable in various fields requiring precise signal analysis, demonstrating robust noise reduction capabilities. As an example, it could encompass security-oriented applications such as radar systems and communication networks.

Parameter Identification of Solid Oxide Fuel Cell Using Elman Neural Network and Dynamic Fitness Distance Balance-Manta Ray Foraging Optimization Algorithm

An accurate solid oxide fuel cell model is a prerequisite for optimizing the operation and state estimation of subsequent cell systems. Hence, this work aimed to utilize a vigoroso algorithmic tool, i.e., Elman neural network, for data prediction to enrich cell measurement data and employ the trained network model for noise reduction of voltage–current data. Furthermore, to obtain reliable cell parameters, a novel parameter identification model based on the dynamic fitness distance balance-manta ray foraging optimization (dFDB-MRFO) algorithm is proposed. Two datasets were applied to extract the electrochemical model and simple electrochemical model parameters of the solid oxide fuel cell model. To verify adequately the superiority of this method, which is compared with another seven conventional heuristic algorithms, four performance indicators were selected as evaluation criteria. Comprehensive case studies demonstrated that through data processing, the precision and robustness of identification could be effectively heightened. In general, the model fitting data obtained via parameter identification using dFDB-MRFO have excellent fitting precision contrast with the measured voltage–current data. Notably, the fitting degree obtained by dFDB-MRFO in the simple electrochemical model reached 99.95% and 99.91% under the two datasets, respectively.

Evaluation of a Vendor-Agnostic Deep Learning Model for Noise Reduction and Image Quality Improvement in Dental CBCT.

To assess the impact of a vendor-agnostic deep learning model (DLM) on image quality parameters and noise reduction in dental cone-beam computed tomography (CBCT) reconstructions. This retrospective study was conducted on CBCT scans of 93 patients (41 males and 52 females, mean age 41.2 years, SD 15.8 years) from a single center using the inclusion criteria of standard radiation dose protocol images. Objective and subjective image quality was assessed in three predefined landmarks through contrast-to-noise ratio (CNR) measurements and visual assessment using a 5-point scale by three experienced readers. The inter-reader reliability and repeatability were calculated. Eighty patients (30 males and 50 females; mean age 41.5 years, SD 15.94 years) were included in this study. The CNR in DLM reconstructions was significantly greater than in native reconstructions, and the mean CNR in regions of interest 1-3 (ROI1-3) in DLM images was 11.12 ± 9.29, while in the case of native reconstructions, it was 7.64 ± 4.33 (p < 0.001). The noise level in native reconstructions was significantly higher than in the DLM reconstructions, and the mean noise level in ROI1-3 in native images was 45.83 ± 25.89, while in the case of DLM reconstructions, it was 35.61 ± 24.28 (p < 0.05). Subjective image quality assessment revealed no statistically significant differences between native and DLM reconstructions. The use of deep learning-based image reconstruction algorithms for CBCT imaging of the oral cavity can improve image quality by enhancing the CNR and lowering the noise.

Steel Wire Rope Damage Width Identification Method Based on Residual Networks and Multi-Channel Feature Fusion

In order to ensure the safety of steel wire rope in various application scenarios, it is particularly important to quantitatively detect the defects of wire rope. Complex detection conditions affect the detection efficiency of wire rope. Therefore, based on the magnetic flux leakage method, this study proposes a method to identify the damage width of steel wire rope for multi-channel fusion of a Hall sensor array. Firstly, the Hall sensor array is used to capture the magnetic flux leakage data of steel wire rope; then, continuous wavelet transform is used to decompose the original data, and moving average filtering is used to denoise each component; the denoised components are merged and converted into a time spectrum, and the time spectrum is classified by ResNet50 image classification model to realize the detection of wire rope damage width. According to the dataset used in this study, the results show that the proposed method performs best in the mainstream noise reduction model; detection accuracy for the width of damage in steel wire ropes is 97%, which proves that the proposed method is effective and feasible.

A genetic programming Rician noise reduction and explainable deep learning model for Alzheimer's diseases severity prediction.

Degradation of magnetic resonance imaging (MRI) remains a challenging issue, with noise being a key damaging component introduced due to a variety of environmental and mechanical factors. The aim of this research work is to addresses the issue of noise reduction and to predict Alzheimer's disease detection efficiently. First, we present a genetic programming (GP) technique for reducing Rician noise in MRI images to pre-process the dataset. To effectively reduce Rician noise, this GP approach combines a Feature Extraction component, GP Optimal Expression, and an Optimum Removal Estimation component. In the second phase, we design and develop an explainable Deep Learning framework. This framework uses a local data-driven interpretation technique based on SHAP values to investigate the relationship between the neural network's estimated AD diagnosis and the input MRI images. In addition, we handle class distribution by combining an oversampling strategy with a minority approach. Several assessment metrics are used to analyze the performance of our proposed model. The proposed method is tested on a variety of medical samples, and the results are compared to those obtained using other comparable approaches. We also test and compare our model to three cutting-edge models: DenseNet169, VGGNet15, and Inceptionv3. The empirical results show that our proposed model outperforms others, particularly in handling basic structures with limited spectral features, lower computational complexity, and less overfitting. This research worked addressed Rician noise issue in MRI images and predict AD severity prediction using explainable deep learning framework.

Research of the impact of noise reduction methods on the quality of audio signal recovery

The subject of the study is the analysis of various filtering algorithms for the quality of the resulting audio files. The importance of audio line filtering has grown significantly in recent years due to its key role in a variety of applications such as speech reduction and artificial intelligence. Taking into account the growing demand for solving problems related to speech recognition, the processing of audio series becomes important for determining the accuracy and efficiency of the obtained solution.The purpose of the work is to study the impact of noise suppression methods on the quality of restoration of an audio signal, which was alternately noisy with one of five types of noise - white, pink, brown, impulse, Gaussian with different power. To achieve the goal, the following tasks were solved: an analysis of the types of noise was carried out and analysis of noise reduction and filtering methods. A generalized model of noise reduction and filtering was developed, and an experiment was planned depending on the type and power of noise. Simulation of the experiment was performed by comparing the parameters of the signal-to-noise ratio before and after the experiment and the peak signal-to-noise ratio in the processed file. The following methods are used: spectral subtraction, filtering based on frequency filters and wavelet transformation.The following results were obtained: depending on the selected noises and algorithms, it was possible to achieve the lowest value of the peak signal-to-noise ratio of 21.52db, and the signal-to-noise ratio increased, which allowed further work with these audio files. The practical significance of this work is the increase in the number of available audio files for further work.Conclusions: the analysis of the obtained results showed that filtering based on frequency filters only worsened the output signal, that is, not only noise, but also useful information is filtered. In all runs, the SNR deteriorates to - 18dB. which is worse than no filtering. Algorithms of spectral subtraction and wavelet transformation improved SNR parameters and output audio files noisy with the most powerful noises in the range of 20dB, which can be considered acceptable for further processing. The results highlight the importance of using denoising and filtering for complex audio processing tasks, particularly neural network training tasks.

Prediction of state-of-health of lithium-ion battery based on CEEMDAN-SG-LSTM combined model

State-of-health (SOH) is an important indicator for the maintenance and safe operation of batteries, and it is crucial for accurately predicting SOH. To address problems that the noise present in the original data lead to inaccurate prediction results. An Long-Short-Term-Memory (LSTM) method for SOH prediction is proposed based on the joint noise reduction model of complete ensemble empirical mode decomposition adaptive noise (CEEDMAN) and Savitzky-Golay (SG) filtering. Firstly, seven health indicators (HIs) were extracted by analyzing the voltage and current curves, and HIs with higher correlation with SOH were selected using Pearson correlation coefficient. Then, Intrinsic Mode Functions (IMF) components generated from SOH by CEEMDAN are divided into noise-component, noise-dominant-component, useful-signal-dominant-component, filtered noise-dominant-component and useful-signal-dominant-component are reconstructed into filtered SOH. Finally, the LSTM model is used for SOH prediction. Experiments show that proposed model captures the capacity regeneration phenomenon well with high prediction accuracy, and errors are all below 1.9%.

Enhancing Low-Light Medical Imaging through Deep Learning-Based Noise Reduction Techniques

Background/ Objectives: Low-light medical imaging is highly challenging in clinical diagnostics due to increased noise levels that mask or obscure important anatomical details. In this respect, conventional noise reduction methods such as Gaussian filtering and median filtering usually lead to a trade-off between noise suppression and the preservation of important features in an image, thus resulting in poor-quality images. More advanced wavelet-based denoising and Non-Local Means methods exhibit superior noise reduction but remain computationally intensive and introduce artifacts. These challenges come with a need to develop more effective and efficient noise-reduction techniques. Methods: This study proposes an end-to-end deep learning framework for low-light medical image enhancement. We present a comprehensive deep-learning framework to enhance low-light medical images by integrating Convolutional Neural Networks with denoising autoencoders to build a robust noise reduction model. The CNN extracts the feature from the noisy input images, while the autoencoder does so for the reconstruction of clean images through the encoding of a noisy input in a lower-dimensional representation for the reduction of noise while retaining critical information. Findings: This study validates the proposed model through rigorous quantitative metrics such as peak signal-to-noise ratio and structural similarity index. These metrics are designed to provide a full assessment of image quality concerning noise reduction capability and preservation of details related to structure. Our model improves traditional methods in PSNR by about 5 dB on average and SSIM by 0.15, which means better noise reduction and preservation of image details. A comparative analysis of traditional techniques for noise reduction has been included, pointing out the advantages of deep learning approaches. Experimental results depict significant improvements over previous approaches. For instance, the proposed model reduces the noise level by up to 40% and facilitates clear and sharp images by up to 30%. In terms of quantification, these improvements manifest in a PSNR value of 35 dB and an SSIM score of 0.85 compared to 30 dB and 0.70 using traditional techniques. Furthermore, the study illustrates the training dynamics, feature maps, and evolution of images to present the model's incremental learning process. Novelty: This study's findings validate the proposed model's efficacy in enhancing diagnosis accuracy and improving patient outcomes in medical imaging. Keywords: Low-light medical imaging, Noise reduction, Convolutional Neural Networks, Denoising autoencoders, Medical diagnostics

Evaluation of Denoising Performance of ResNet Deep Learning Model for Ultrasound Images Corresponding to Two Frequency Parameters.

Ultrasound imaging is widely used for accurate diagnosis due to its noninvasive nature and the absence of radiation exposure, which is achieved by controlling the scan frequency. In addition, Gaussian and speckle noises degrade image quality. To address this issue, filtering techniques are typically used in the spatial domain. Recently, deep learning models have been increasingly applied in the field of medical imaging. In this study, we evaluated the effectiveness of a convolutional neural network-based residual network (ResNet) deep learning model for noise reduction when Gaussian and speckle noises were present. We compared the results with those obtained from conventional filtering techniques. A dataset of 500 images was prepared, and Gaussian and speckle noises were added to create noisy input images. The dataset was divided into training, validation, and test sets in an 8:1:1 ratio. The ResNet deep learning model, comprising 16 residual blocks, was trained using optimized hyperparameters, including the learning rate, optimization function, and loss function. For quantitative analysis, we calculated the normalized noise power spectrum, peak signal-to-noise ratio, and root mean square error. Our findings showed that the ResNet deep learning model exhibited superior noise reduction performance to median, Wiener, and median-modified Wiener filter algorithms.

The Impact of AI on Metal Artifacts in CBCT Oral Cavity Imaging.

This study aimed to assess the impact of artificial intelligence (AI)-driven noise reduction algorithms on metal artifacts and image quality parameters in cone-beam computed tomography (CBCT) images of the oral cavity. This retrospective study included 70 patients, 61 of whom were analyzed after excluding those with severe motion artifacts. CBCT scans, performed using a Hyperion X9 PRO 13 × 10 CBCT machine, included images with dental implants, amalgam fillings, orthodontic appliances, root canal fillings, and crowns. Images were processed with the ClariCT.AI deep learning model (DLM) for noise reduction. Objective image quality was assessed using metrics such as the differentiation between voxel values (ΔVVs), the artifact index (AIx), and the contrast-to-noise ratio (CNR). Subjective assessments were performed by two experienced readers, who rated overall image quality and artifact intensity on predefined scales. Compared with native images, DLM reconstructions significantly reduced the AIx and increased the CNR (p < 0.001), indicating improved image clarity and artifact reduction. Subjective assessments also favored DLM images, with higher ratings for overall image quality and lower artifact intensity (p < 0.001). However, the ΔVV values were similar between the native and DLM images, indicating that while the DLM reduced noise, it maintained the overall density distribution. Orthodontic appliances produced the most pronounced artifacts, while implants generated the least. AI-based noise reduction using ClariCT.AI significantly enhances CBCT image quality by reducing noise and metal artifacts, thereby improving diagnostic accuracy and treatment planning. Further research with larger, multicenter cohorts is recommended to validate these findings.

Enabling Low-Dose In Vivo Benchtop X-ray Fluorescence Computed Tomography through Deep-Learning-Based Denoising.

X-ray Fluorescence Computed Tomography (XFCT) is an emerging non-invasive imaging technique providing high-resolution molecular-level data. However, increased sensitivity with current benchtop X-ray sources comes at the cost of high radiation exposure. Artificial Intelligence (AI), particularly deep learning (DL), has revolutionized medical imaging by delivering high-quality images in the presence of noise. In XFCT, traditional methods rely on complex algorithms for background noise reduction, but AI holds promise in addressing high-dose concerns. We present an optimized Swin-Conv-UNet (SCUNet) model for background noise reduction in X-ray fluorescence (XRF) images at low tracer concentrations. Our method's effectiveness is evaluated against higher-dose images, while various denoising techniques exist for X-ray and computed tomography (CT) techniques, only a few address XFCT. The DL model is trained and assessed using augmented data, focusing on background noise reduction. Image quality is measured using peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM), comparing outcomes with 100% X-ray-dose images. Results demonstrate that the proposed algorithm yields high-quality images from low-dose inputs, with maximum PSNR of 39.05 and SSIM of 0.86. The model outperforms block-matching and 3D filtering (BM3D), block-matching and 4D filtering (BM4D), non-local means (NLM), denoising convolutional neural network (DnCNN), and SCUNet in both visual inspection and quantitative analysis, particularly in high-noise scenarios. This indicates the potential of AI, specifically the SCUNet model, in significantly improving XFCT imaging by mitigating the trade-off between sensitivity and radiation exposure.

Numerical model for noise reduction of small vertical-axis wind turbines

Abstract. Small vertical-axis wind turbines are a promising solution for affordable and clean energy, but their noise emissions present a challenge to public acceptance. Numerous blade designs have been aimed at reducing noise but often come with a decrease in wind turbine aerodynamic efficiency. In this study, the acoustic power and torque of a 5 kW vertical-axis wind turbine (VAWT) were simulated by using different mesh sizes and turbulence models. The simulated torque and noise of the turbine have significant sensitivity to the mesh size, so suitable mesh sizes were determined for the near-wall and rotating regions that can be used as a design reference for future turbines with similar operating conditions. The selection of the turbulence model was found to affect the predicted torque by about 10 % and the predicted tip noise by about 2 dB. The selected mesh size and turbulence model were then applied to simulating the effectiveness of three common noise mitigation techniques: a mask, deflector, and wall roughness. The results showed that deflectors are suitable for noise reduction of small VAWTs. This paper provides valuable information on simulating noise propagation from small VAWTs and the optimal noise reduction techniques.

Research on U-Net seismic signal denoising combined with residual dense blocks

Aiming to address the limited noise reduction capabilities of conventional methods for reducing noise in seismic signals, the paper proposes a noise reduction model based on the RDBU-Net network. This model utilizes a residual dense block (RDB) instead of conventional convolutional layers in the U-Net network to enhance the feature extraction capacity for same-band noise, thereby elevating the signal-to-noise ratio (SNR) of seismic signals. The RDBU-Net model is trained, validated, and tested using the global seismic dataset from Stanford University. In comparison with the wavelet threshold method, the denoising RDB model, and the U-Net model, the RDBU-Net model demonstrates an improvement in SNRs by 7.82 dB, 6.13 dB, and 2.9 dB, respectively. Additionally, the root mean square errors are reduced by 0.4812, 0.3736, and 0.1938, and the correlation coefficients are enhanced by 0.3818, 0.2714, and 0.1205. The RDBU-Net model proposed in this study effectively enhances the SNR of seismic signals and offers fresh insights into eliminating noise within the same frequency band of seismic signals.

Speech enhancement augmentation for robust speech recognition in noisy environments

Abstract. The use of augmentations as a data enrichment method has become an important element in improving the performance of speech recognition systems. To work effectively in noisy conditions, augmentation is usually used to simulate the presence of background noise. However, the quality of speech recognition on samples pre-processed by noise reduction models does not increase. This paper proposes a new approach to speech data augmentation when training ASR systems, intended for their joint use with models for speech enhancement. It was based on the creation of several additional data samples containing speech samples processed by the enhancement model. The proposed approach was tested on the E-Branchformer neural network model using data from the Librispeech set. The quality of speech samples was assessed using the DNSMOS metric. By means of a 100-hour sample of clean speech samples it was shown that the proposed augmentation allows for an improvement in the WER metric of more than 4% in absolute value compared to the generally accepted approach based on adding noisy speech samples. Experiments on 960-hour data demonstrated the robustness of this approach as the training set size increased.

Improving transmission loss performance of aluminum extrusion in railway vehicles using lightweight material

This study developed a method for improving the transmission loss performance of aluminum extrusion by reinforcing the inner space of the extruded material with a lightweight material, such as acrylic plastic. First, the dynamic characteristics of the aluminum extrusion were analyzed, and the characteristics of noise transmission were evaluated through an acoustic mode analysis of the air layer for the inner space of the extruded material. Furthermore, to improve the transmission loss performance of the aluminum extrusion, methods were developed for increasing the vibration stiffness and altering the acoustic mode of the inner space. Additionally, a simulation analysis of the vibration mode and a response analysis based on the excitation of sound were performed to validate the proposed model; the effectiveness of the model for noise and vibration reduction was compared with that of the existing model. This analysis was conducted using models wherein the acrylic plastic was partially and totally inserted into the interior space of aluminum extrusion. The results indicated that the noise performance of the short reinforcement model with the partially inserted acrylic plastic was improved by 3 dB and that of the long reinforcement model with the totally inserted acrylic plastic was degraded by &gt; 4 dB compared with the existing aluminum panel.

ADDITION: Detecting Adversarial Examples With Image-Dependent Noise Reduction

Notwithstanding the tremendous success of deep neural networks in a range of realms, previous studies have shown that these learning models are exposed to an inherent hazard called <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">adversarial example</i> — images to which an elaborate perturbation is maliciously added could deceive a network, which entails the study of countermeasures urgently. However, existing solutions suffer from some weaknesses, e.g. parameters are usually determined empirically in some processing-based detection methods might result in a sub-optimal effect, and the directly performed processing on images might affect the classification of benign samples, leading to increment of false positive. In this paper, we propose a novel imAge-DepenDent noIse reducTION (ADDITION) model based on deep learning for adversarial detection. The ADDITION model can adaptively convert the adversarial perturbation in each image to approximate Gaussian noise by injecting image-dependent additional noise, then perform noise reduction to eliminate the adversarial perturbation, and finally detect adversarial examples by examining the classification inconsistency between the input image and its denoised version. The ADDITION model is trained end-to-end on benign samples without any prior knowledge of adversarial attacks, and thus avoid time-consuming task of generating adversarial examples in practical use. We generate more than 220,000 adversarial examples based on six attack algorithms for evaluation and present state-of-the-art comparisons on three real-word datasets. Extensive experiments demonstrate that our proposed method achieves improved performance in both detection accuracy rate and false positive rate.

Research on Digital Virtual Interaction Technology for Intelligent Image Recognition in Woodcarving Protection

Abstract Advances in AI and VR have enhanced digital virtual interaction, crucial for preserving wood carving art. This study introduces an optimized system using intelligent image recognition, overcoming challenges in current systems. It involves image collection, noise reduction, feature extraction, normalization, and transfer learning-based model training. The system achieved a 93.34% recognition accuracy, surpassing traditional methods and improving overall performance.

Speech Enhancement Using Joint DNN‐NMF Model Learned with Multi‐Objective Frequency Differential Spectrum Loss Function

We propose a multi‐objective joint model of non‐negative matrix factorization (NMF) and deep neural network (DNN) with a new loss function for speech enhancement. The proposed loss function (LMOFD) is a weighted combination of a frequency differential spectrum mean squared error (MSE)‐based loss function (LFD) and a multi‐objective MSE loss function (LMO). The conventional MSE loss function computes the discrepancy between the estimated speech and clean speech across all frequencies, disregarding the process of changing amplitude in the frequency domain which contains valuable information. The differential spectrum representation retains spectral peaks that carry important information. Using this representation helps to ensure that this information in the speech signal is reserved. Also, on the other hand, noise spectra typically have a flat shape and as the differential operation makes the flat spectral partly close to zero, the differential spectrum is resistant to noises with smooth structures. Thus, we propose using a frequency‐differentiated loss function that considers the magnitude spectrum differentiations between the neighboring frequency bins in each time frame. This approach maintains the spectrum variations of the objective signal in the frequency domain, which can effectively reduce the noise deterioration effects. The multi‐objective MSE term (LMO) is a combined two‐loss function related to the NMF coefficients which are the intermediate output targets, and the original spectral signals as the actual output targets. The use of encoded NMF coefficients as low‐dimensional structural features for DNN serves as prior knowledge and helps the learning process. LMO is used beside LFD to take advantage of both the properties of the original and the differential spectrum in the training loss function. Moreover, a DNN‐based noise classification and fusion strategy (NCF) is proposed to exploit a discriminative model for noise reduction. The experiments reveal the improvements of the proposed approach compared to the previous methods.

Heart rate detection of ballistocardiogram based on improved DAE and template matching method

Ballistocardiogram (BCG) is a non-invasive physiological signal detection method that can be used for non-contact detection of resting heart rate (RHR) and has been widely used in human health monitoring. However, the BCG signal is vulnerable to noise, making it challenging to accurately measure heart rate. In this paper, we propose a noise reduction model for the BCG signal based on Gramian Angular Field (GAF) and an improved Denoising Autoencoder (DAE), referred to as the GDAE model, to accurately detect heart rate in noise-contaminated signals. First, the Gramian Angular Field transform is used to convert the one-dimensional BCG signal into two-dimensional image information, highlighting the difference between the signal centroid information and noise information; after that, the transformed image is denoised by the Denoising Autoencoder to obtain a denoised BCG signal. After noise reduction, the heart rate of the subject is calculated using the adaptive template matching method. The test proves that under the strong noise interference, the proposed method improves the recall by 6.87% and the accuracy by 6.02% compared with the traditional method, indicating a better detection effect. Furthermore, the comparison test shows that the GDAE model has a significant noise reduction effect on the BCG signal, which improves the practicality of the BCG method for heart rate detection.

A spatially variant high-order variational model for Rician noise removal.

Rician noise removal is an important problem in magnetic resonance (MR) imaging. Among the existing approaches, the variational method is an essential mathematical technique for Rician noise reduction. The previous variational methods mainly employ the total variation (TV) regularizer, which is a first-order term. Although the TV regularizer is able to remove noise while preserving object edges, it suffers the staircase effect. Besides, the adaptability has received little research attention. To this end, we propose a spatially variant high-order variational model (SVHOVM) for Rician noise reduction. We introduce a spatially variant TV regularizer, which can adjust the smoothing strength for each pixel depending on its characteristics. Furthermore, SVHOVM utilizes the bounded Hessian (BH) regularizer to diminish the staircase effect generated by the TV term. We develop a split Bregman algorithm to solve the proposed minimization problem. Extensive experiments are performed to demonstrate the superiority of SVHOVM over some existing variational models for Rician noise removal.