Proposed Noise Reduction Method Research Articles

An ECG denoising technique based on AHIN block and gradient difference max loss

The electrocardiogram (ECG) signal is susceptible to interference from unknown noises during the acquisition process due to their low frequency and amplitude, resulting in the loss of significant information in the signals. Recent advancements in deep learning models have shown promising results in signal processing. However, these models lack robustness against various types of noise and often overlook the gradient difference between denoised and original signals. In this study, we propose a novel deep learning denoising method based on an attention half instance normalization block (AHIN block) and a gradient difference max loss function (GDM Loss). Our approach consists of two stages: firstly, we input the noisy ECG signal to obtain a denoised version; secondly, we reconstruct the denoised signal by fusing preliminary results from the first stage while correcting waveform distortions caused by initial denoising to minimize information loss. Additionally, we introduce a new loss function that considers differences between slopes of the denoised ECG signal and clean ECG signal. To validate our proposed method's effectiveness, extensive experiments were conducted on both our model architecture and loss function compared with other state-of-the-art methods. Results demonstrate that our approach achieves excellent performance in terms of both signal-to-noise ratio (SNR) and root-mean-square error (RMSE). The proposed noise reduction method improves 8.86%, 12.05% and 10.50% respectively under BW, MA and EM noise.

Automatic bioacoustics noise reduction method based on a deep feature loss network

Acoustic sensors that collect acoustic data over extended periods and broad ranges are widely used in bioacoustics monitoring. However, in open environments, acoustic data collected using acoustic sensors can be subject to interference from various real-world noises, thereby influencing the subsequent analysis and processing of bioacoustic data. Existing bioacoustic noise reduction methods are limited in their application because of their low efficiency, unsuitability for non-stationary noise, generally unimproved signal-to-noise ratio (SNR) efficacy, and considerable amounts of residual noise. These limitations hinder the effective processing of recorded signals for which extraneous noise overlaps with bird vocalizations. In this study, we propose a bioacoustic noise reduction method based on a deep feature loss network for bird sounds. The method has a rapid denoising speed and can more effectively remove background noise from field recording signals without distorting the bird acoustic spectrum. The denoising effects of the proposed method were compared with those of a speech enhancement generative adversarial network, web real-time communications denoising, and other noise reduction methods. The denoising ability of these methods for different noises was evaluated using spectrograms and objective evaluation measures such as the SNR and perceptual evaluation of speech quality (PESQ). The experimental results revealed that our proposed noise reduction method can obtain higher SNRs and PESQ scores than other noise reduction methods, with the SNR increasing by up to 35.83 dB following denoising.

A novel method for noise reduction of blade tip timing signals based on sparse representation and dictionary learning

Blade Tip Timing Technology (BTT) is a new type of blade vibration measurement method with non-contact measurement capabilities, which is associated with high efficiency and convenience. However, due to under-sampling feature of the BTT signal, the method can only measure part of the vibration information. Therefore, various BTT spectral reconstruction algorithms have been developed based on the under-sampling feature and sparsity of the blade vibration spectrum. Because of the harsh working environment of the blades, BTT signal typically contains high-level Gaussian noise in actual measurement, which significantly reduces frequency-domain sparsity and even aliases the vibration modals, resulting in a decrease in the performance of sparse reconstruction algorithms. In this paper, a multi-sequences BTT signal noise reduction method is proposed, which is based on sparse representation to suppress Gaussian noise. Through appropriately modifying the constraint in the ℓ0 optimization problem, the method filters out the noise elements in the sparse vector of the BTT signal. Then, denoising is completed by sparse inverse transformation. Finally, a global average-based singular value decomposition dictionary learning algorithm (GA-K-SVD) is proposed to generate an over-complete sparse dictionary which is adaptable to the original signal, to increase the effectiveness of GA-K-SVD. At last, simulation and experience are carried out to verify the performance optimization effect for sparse reconstruction algorithm and the effectiveness of proposed noise reduction method.

Enhancement and labelling of OCT images

Abstract Optical Coherence Tomography (OCT) is a noninvasive technique for visualizing retinal cross-sections, assisting in diagnosing and monitoring retinal diseases. This study presents an effective OCT image noise reduction and segmentation method. Due to the high resolution of our primary image database, a visual comparison of filter performance was challenging. To address this, we utilized a high-noise image dataset from Duke University, enabling a better evaluation of noise reduction filters. Our method was compared against eight widely-used image filters, including Gaussian, Low pass,Wavelet domain filtering, Lee filter, Anisotropic diffusion, Bilateral filter, Total variational filter, and BM3D filter. Both visual and quantitative analyses were conducted using no-reference performance parameters, namely Gradient Magnitude Similarity Deviation (GMSD), Standard Deviation of Wavelet Coefficients (SDWC), Focus Measure with Tengrade Variance (FMTV), Perception-based Image Quality Evaluator (PIQE), and Visual Information Fidelity (VIF). The results demonstrated the superiority of our proposed filter in terms of noise reduction performance while maintaining the sharpness of retinal layers. Quantitative analysis revealed notable performance gains, including improvements of 63.27% to 83.24% with GMSD, consistent edge strength enhancement of 9.4% to 9.97% using SDWC, gains in image quality between 51.99% and 54.64% with FMTV, performance improvements ranging from 7.25% to 23.97% in terms of PIQE, and a substantial increase in performance varying from 16.31% to 27.69% when assessing the impact on retinal layer quality using VIF. Using the Kruskal-Wallis test, our proposed noise reduction method shows statistical significance for all quantitative parameters. Additionally, our clustering algorithm effectively separated the foreground, including retinal layers and vitreous detachment, from the background and identified an area representing the region between retinal layers where fluid accumulates. We’ve successfully achieved OCT image enhancement, along with distinct foreground and background segmentation.

Noise Reduction of OCT Images Based on the External Patch Prior Guided Internal Clustering and Morphological Analysis

Optical coherence tomography (OCT) is widely used in biomedical imaging. However, noise severely affects diagnosing and identifying diseased tissues on OCT images. Here, a noise reduction method based on the external patch prior guided internal clustering and morphological analysis (E2PGICMA) is developed to remove the noise of OCT images. The external patch prior guided internal clustering algorithm is used to reduce speckle noise. The morphological analysis algorithm is employed to the background for contrast enhancement. OCT images of in vivo normal skin tissues were analyzed to remove noise using the proposed method. The estimated standard deviations of the noise were chosen as different values for evaluating the quantitative metrics. The visual quality improvement includes more textures and fine detail preservation. The denoising effects of different methods were compared. Then, quantitative and qualitative evaluations of this proposed method were conducted. The results demonstrated that the SNR, PSNR, and XCOR were higher than those of the other noise-reduction methods, reaching 15.05 dB, 27.48 dB, and 0.9959, respectively. Furthermore, the presented method’s noise reduction ratio (NRR) reached 0.8999. This proposed method can efficiently remove the background and speckle noise. Improving the proposed noise reduction method would outperform existing state-of-the-art OCT despeckling methods.

Classification of Children's Heart Sounds With Noise Reduction Based on Variational Modal Decomposition.

PurposeChildren's heart sounds were denoised to improve the performance of the intelligent diagnosis.MethodsA combined noise reduction method based on variational modal decomposition (VMD) and wavelet soft threshold algorithm (WST) was proposed, and used to denoise 103 phonocardiogram samples. Features were extracted after denoising and employed for an intelligent diagnosis model to verify the effect of the denoising method.ResultsThe noise in children's phonocardiograms, especially crying noise, was suppressed. The signal-to-noise ratio obtained by the method for normal heart sounds was 14.69 dB at 5 dB Gaussian noise, which was higher than that obtained by WST only and the other VMD denoising method. Intelligent classification showed that the accuracy, sensitivity and specificity of the classification system for congenital heart diseases were 92.23, 92.42, and 91.89%, respectively and better than those with WST only.ConclusionThe proposed noise reduction method effectively eliminates noise in children's phonocardiograms and improves the performance of intelligent screening for the children with congenital heart diseases.

Improved Wavelet Packet Noise Reduction for Microseismic Data via Fuzzy Partition

It is crucial to retain as many signal components as possible, while noise is eliminated for wavelet packet noise reduction algorithms. Conventional thresholding functions take coefficients smaller than a given threshold as the noise component and get them removed in many ways. In this letter, considering the potential probability of being the signal-related component of coefficients whose value is smaller than the given threshold and the nonnegligible fact that microseismic (MS) event is sparse compared with the noise, we proposed a new wavelet packet-based denoising method via fuzzy partition. First, instead of a hard partition from a given threshold, wavelet packet coefficients get reduced by a fuzzy partition to retain more potential signal elements and suppress the noise. Second, we employ fuzzy c-means (FCM) clustering to identify the interval period of the MS event further to remove the residual noise and additional resonance in processed time series. We tested our method on synthetic datasets and real-field data from an MS monitoring experiment in a coal mine in Sichuan Basin, China. We utilize Pearson correlation coefficient and root-mean-square error between ideal signal and denoised data as performance indicators in synthetic tests, while sample entropy and kurtosis of denoised data are involved in the real field dataset. Test results from synthetic datasets and the real field dataset demonstrate that the proposed noise reduction method is superior to traditional hard-, soft-, and garrote-thresholding and is more applicable and effective in MS data processing.

High-Quality Signal Denoising and Deep Feature Representation and Its Application to the ENPEMF

The Earth’s natural pulsed electromagnetic field (ENPEMF) signal is released by the instantaneous disturbance of the Earth’s naturally varying magnetic field, contains a large amount of information about the changing geological structures and their kinetic principles. The ENPEMF signals acquired by the equipment are often accompanied by a great deal of noise, and effective noise reduction is required for accurate signal analysis. We propose a new noise reduction method for a very low signal-to-noise ratio (SNR), called the Blaschke unwinding adaptive Fourier decomposition, and use synthetic signals to demonstrate the high accuracy of the proposed noise reduction method. Meanwhile, the Root-phase deep feature representation (RPDP) of the ENPEMF signal acquired during the Lushan Ms7.0 earthquake is explored using the roots obtained by Blaschke unwinding. The results show that the time-based RPDP can accurately reveal the deep feature of the ENPEMF signal before and after the earthquake, and extract the anomalous changes, potentially revealing the electromagnetic anomaly information contained in the ENPEMF signal.

The noise reduction of medical radiography images using fractional moments

Background: This paper presents a method to improve medical radiography images based on the use of statistical signal moments. Methods: In this paper, the image with noise is considered as a statistical signal, and the noise reduction is performed by using fractional moments. The fractional moment’s method, on the one hand, has a speed similar to the moment method, and, on the other hand, has not the limitations of the moment method, which sometimes achieves inaccurate results. The proposed method is ultimately examined on radiographic images (CT). Results: The information obtained from the fractional moments of the received signal is a criterion to estimate the noise parameters and the gray scales of the main image. One of the limitations of the proposed method is that the image should be sent several times, because in statistical discussions, we cannot make a decision with only one sample. The error of the proposed noise reduction method in terms of the number of times the original image was sent, is about 0.009, 0.0009, 0.0002, and 0.0001, for n = 3, n = 6, n = 9 and n = 14, respectively. Conclusion: The simulation results show that the proposed method is more effective than the most conventional noise reduction methods, both in the low signal to noise ratio and in terms of image quality, and is more powerful than the most notable noise removal methods in restoring the subtleties and image details.

Novel noise reduction method based on improved empirical wavelet transform and kurtosis for partial discharge signal of high‐voltage cables

Partial discharge (PD) detection is useful to the online monitoring of high-voltage cables, which are affected by periodic narrowband interference and random white noise. To suppress the influence of complex noise signals on the measurement of PD signals, a novel noise reduction method based on improved empirical wavelet transform (IEWT) and kurtosis for the PD signals of high-voltage cables is proposed. First, empirical wavelet function (EWF) signals arranged in order of frequency are obtained by decomposing the signal with IEWT. Then, the kurtosis criterion is introduced, and the reconstructed inherent modal function is adaptively screened. Finally, the improved threshold function is used in eliminating residual noise information in the reconstructed signal. Compared with the noise reduction methods based on EWT and EMD, the proposed noise reduction method can effectively suppress noise information in PD and has good practicability.

Safety state identification of concrete pumping pipeline based on multi-channel audio signals

Based on the principle of superposition and attenuation of sound propagation, a Multi-point Multi-channel noise Reduction Method (MMRM) for audio monitoring is proposed. With the help of the proposed noise reduction method, the sounds made by the concrete pumping-pipelines themselves on the construction site and the environmental noises are separated in real time. Then, the Mel-Frequency Cepstral Coefficients (MFCCs) and spectral centroids of the filtered pumping-pipeline signals are extracted in real time as the features. Finally, the distance between the currently obtained feature vector of the pipeline signals and the previous accumulated MFCC clustering are determined as the parameter for safety identification. In order to verify the actual effects of the MMRM and the analysis accuracy of MFCCs in the clustering, several numerical tests and field measurements are carried out. The numerical results show that the MMRM has an excellent performance on noise reduction; the feature vectors composed of the MFCCs with length no greater than 4 are sufficient for the recognition and clustering of pumping-pipeline operation state.

Diagnozowanie uszkodzeń wielostopniowej pompy odśrodkowej z wykorzystaniem metody odszumiania sygnału drgań w oparciu o średnie nielokalne

In real industry environment, the signal characteristics of multistage centrifugal pump vibration signal are easily submerged by strong background noise. To settle this problem, the nonlocal means (NLM) approach is proposed for the denoising of multistage centrifugal pump in this paper. Utilizing the similarity theory, the NLM method has achieved a wide range of applications in the fields of image processing and biomedical signal denoising. Due to the periodic characteristics and redundancy, NLM is successfully applied to the de-noising of 1-D machinery vibration signal. The numerical simulation experiments with different SNRs verify the effectiveness and the superiority of the proposed method. Besides, the selection principles of core parameters in NLM are discussed. The real engineering cases analysis demonstrates that the NLM can effectively filter out the background noise and realize the weak fault feature enhancements. The proposed noise reduction method is superior to traditional wavelet coefficient method.

Design of an induction machine with damper windings for noise and vibrations reduction

PurposeThe purpose of this paper is to present a design method for induction machines including a three-phase damper winding for noise and vibrations reduction.Design/methodology/approachIn the first part, the principle of the damper winding is recalled. The second part presents the iterative design method which is applied on a 4-kW pulse width modulation (PWM)-fed induction machine to study the impact of the additional winding on the geometry. In the third part, the finite-element method is used to validate the designed geometry and highlight the harmonic flux density reduction. Finally, some experimental results are given.FindingsThe study shows that the impact of the additional three-phase winding on the geometry and weight of the machine is low. Moreover, the proposed noise reduction method allows one to reduce the total noise level of a PWM-fed induction machine up to 8.5 dBA.Originality/valueThe originality of the paper concerns the design and characterization of a three-phase damper winding for a noiseless induction machine. The principle of this proposed noise reduction method is new and has been patented.

White noise reduction for wideband linear array signal processing

The performance of wideband array signal processing algorithms is dependant on the noise level in the system. In this thesis, a method is proposed for reducing the level of white noise in wideband arrays via a judiciously designed spatial transformation followed by a bank of high-pass filters. The method is initially introduced for uniform linear arrays (ULAs) and analysed in detail. The spectrum of the signal and noise after being processed by the proposed noise reduction method is analysed, and the correlation matrix of the processed noise is derived. The reduced noise level leads to a higher signal-to-noise ratio (SNR) for the system, which can have a significant effect on the performance improvement of various beamforming methods and other array signal processing applications such as direction of arrival (DOA) estimation. The performance of two well-known beamformers, the reference signal based (RSB) beamformer and the linearly constrained minimum variance (LCMV) beamformer is reviewed. Then, the theoretical effect of applying the proposed noise reduction method as a pre-processing step on the performance enhancement of RSB and LCMV beamformers is studied. The theoretical results are then confirmed by simulation. As a representative example of wideband DOA estimation application, a compressive sensing-based DOA estimation method is employed to demonstrate the improved estimation by applying the pre-processing noise reduction method, which is confirmed by simulation. Next, the idea is extended to wideband non-uniform linear arrays (NLAs). Since, NLA does not have a uniform spacing, the beam response of the row vectors of the transformation is distorted. Therefore, the transformation is re-designed using the least squares method to satisfy the band-pass requirements of the transformation. Simulation results show a satisfactory improvement in the the performance of RSB and LCMV beamformers for the NLA structure. The idea is further extended to uniform rectangular arrays (URAs) and uniform circular arrays (UCAs), as two major types of the planar arrays. Two methods are proposed for reducing the effect of white noise in wideband URAs and for each one, a different transformation is designed. The first one is based on a two-dimensional (2D) transformation and the second one is an adaptation of the method developed for the ULA case. The developed method for the UCA structure is based on a one-dimensional (1D) transformation, with modified modulation for the transformation to satisfy the required band-pass characteristics of the transformation. Same as linear array structures, the RSB and LCMV beamformers are used to demonstrate the performance enhancement of the method for planar arrays.

Noise Reduction of Steel Cord Conveyor Belt Defect Electromagnetic Signal by Combined Use of Improved Wavelet and EMD

In order to reduce the noise of a defect electromagnetic signal of the steel cord conveyor belt used in coal mines, a new signal noise reduction method by combined use of the improved threshold wavelet and Empirical Mode Decomposition (EMD) is proposed. Firstly, the denoising method based on the improved threshold wavelet is applied to reduce the noise of a defect electromagnetic signal obtained by an electromagnetic testing system. Then, the EMD is used to decompose the denoised signal and then the effective Intrinsic Mode Function (IMF) is extracted by the dominant eigenvalue strategy. Finally, the signal reconstruction is carried out by utilizing the obtained IMF. In order to verify the proposed noise reduction method, the experiments are carried out in two cases including the defective joint and steel wire rope break. The experimental results show that the proposed method in this paper obtains the higher Signal to Noise Ratio (SNR) for the defect electromagnetic signal noise reduction of steel cord conveyor belts.

Dual-Microphone Noise Reduction in Car Environments With Determinant Analysis of Input Correlation Matrix

In this paper, a novel noise reduction technique is proposed to improve the speech interface performance in car environments. The proposed noise reduction method with dual microphones is primarily based on the determinant analysis of the input correlation matrix. Through the analysis, a robust feature for speech activity detection and signal-to-noise ratio (SNR) estimation is derived. Using the feature, the SNR of each time-frequency component is estimated, and an enhanced speech signal is obtained through Wiener filtering. To evaluate the proposed noise reduction technique, we constructed a database in a real car environment and comparatively analyzed the performances of noise reduction methods. The results show that meaningful SNR and perceptual speech quality improvements with less signal distortion are achieved compared with the other competing methods.

A robust noise reduction technique for time resolved CT.

To develop a noise reduction method for time resolved CT data, especially those with significant patient motion. PArtial TEmporal Nonlocal (PATEN) means is a technique that uses the redundant information in time-resolved CT data to achieve noise reduction. In this method, partial temporal profiles are used to determine the similarity (or weight) between pixels, and the similarity search makes use of both spatial and temporal information, providing robustness to patient motion. The performance of the PATEN filter was qualitatively and quantitatively evaluated with nine cardiac CT patient data sets and five CT brain perfusion patient data sets. In cardiac CT, PATEN was applied to reduce noise primarily in the reduced-dose phases created with electrocardiographic (ECG) pulsing. CT number accuracy and noise reduction were evaluated in both full-dose phases and reduced-dose phases between filtered backprojection images and PATEN filtered images. In CT brain perfusion, simulated quarter dose data were obtained by adding noise to the raw data of a routine dose scan. PATEN was applied to the simulated low-dose images. Image noise, time-intensity profile accuracy, and perfusion parameter maps were compared among low-dose, low-dose+PATEN filter, and full-dose images. The noise reduction performance of PATEN was compared to a previously proposed noise reduction method, time-intensity profile similarity (TIPS) bilateral filtering. In 4D cardiac CT, after PATEN filtering, the image noise in the reduced-dose phases was greatly reduced, making anatomical structures easier to identify. The mean decreases in noise values between the original and PATEN images were 11.0% and 53.8% for the full and reduced-dose phases of the cardiac cycle, respectively. TIPS could not achieve effective noise reduction. In CT brain perfusion, PATEN achieved a 55.8%-66.3% decrease in image noise in the low-dose images. The contrast to noise ratio in the axial images was increased and was comparable to the full-dose images. Differentiation of anatomical structure in the PATEN images and corresponding quantitative perfusion parameter maps were preferred by two neuroradiologists compared to the simulated low-dose and TIPS results. The mean perfusion parameters calculated from the PATEN images agreed with those determined from full-dose data to within 12% and 20% for normal and diseased regions. In ECG-gated cardiac CT, where the dose had already been reduced by a factor of 5 in the reduced-dose phases, PATEN achieved a 53.8% noise reduction, which decreased the noise level in the reduced-dose phases close to that of the full-dose phases. In CT brain perfusion, a fourfold dose reduction was demonstrated to be achievable by PATEN filtering, which improved quantitative perfusion analysis. PATEN can be used to effectively reduce image noise to improve image quality, even when significant motion occurred between temporal samples.

A Hybrid Technique of Noise Reduction with Periductal Fibrosis Ultrasound Images for Periductal Fibrosis Detection System of Cholangiocarcinoma Surveillance

The Cholangiocarcinoma (CCA) is a serious public health problem. The Periductal fibrosis (PDF) ultrasound images are applied for CCA surveillance because it is no side effect of radiation with patients, easy to portability and low cost. In contrast, the common problem of ultrasound images are speckle noise in which decreases the PDF detection performance. In this paper proposes a hybrid noise reduction method in the PDF detection system. The proposed noise reduction method by applying the Median filter and Fast Fourier transform based on PDF ultrasound images. The experimental results give the best performance for PDF detection system. A success rate of proposed method achieved at 70.89%.

Controlling Tradeoff Between Approximation Accuracy and Complexity of a Smooth Function in a Reproducing Kernel Hilbert Space for Noise Reduction

Noise reduction algorithms are widely used to mitigate noise effects on speech to improve the robustness of speech technology applications. However, they inevitably cause speech distortion. The tradeoff between noise reduction and speech distortion is a key concern in designing noise reduction algorithms. This study proposes a novel framework for noise reduction by considering this tradeoff. We regard speech estimation as a function approximation problem in a regularized reproducing kernel Hilbert space (RKHS). In the estimation, the objective function is formulated to find an approximation function by controlling the tradeoff between approximation accuracy and function complexity. For noisy observations, this is equivalent to controlling the tradeoff between noise reduction and speech distortion. Since the target function is approximated in an RKHS, either a linear or nonlinear mapping function can be naturally incorporated in the estimation by a “kernel trick”. Traditional signal subspace and Wiener filtering based noise reduction can be derived as special cases when a linear kernel function is applied in this framework. We first provided a theoretical analysis of the tradeoff property of the framework in noise reduction. Then we applied our proposed noise reduction method in speech enhancement and noisy robust speech recognition experiments. Compared to several classical noise reduction methods, our proposed method showed promising advantages.

A modified weighted overlap and add-based spectral subtraction method

The Weighted Overlap and Add (WOLA) method is an optimization of frequency modulated DFT filter banks. However, the decrement in frequency resolution of speech signal caused by WOLA leads to noise estimation error. In this study, a speech power compression and gain compensation method was proposed to solve the low-resolution problem from the WOLA process. Finally, a multi-band spectral subtraction method was developed by combining the WOLA and the spectral subtraction algorithm in frequency domain. The noise reduction degrees were determined by the individual Signal-to-Noise Ratios (SNRs) in each frequency bands. Objective evaluation of the proposed noise reduction method was performed based on Itakura-Saito Distortion Measure under different types of noise. Statistical results showed that this proposed method could improve noisy speech by 5 dB SNR with little distortion. This noise reduction method has an advantage on low computation load, real-time processing and good performance on noise reduction. So this method can be implemented in embedded devices, e.g. hearing aid and cochlear implants. # This work is supported by National Natural Science Foundation of China (11104316). and Shanghai Natural Science Foundation (11ZR1446000).