Denoising Stage Research Articles

Time-Frequency Analysis and Neural Networks for Detecting Short-Circuited Turns in Transformers in Both Transient and Steady-State Regimes Using Vibration Signals

Transformers are vital elements in electrical networks, but they are prone to various faults throughout their service life. Among these, a winding short-circuit fault is of particular concern to researchers, as it is a crucial and vulnerable component of the transformers. Therefore, if this fault is not addressed at an early stage, it can increase costs for users and affect industrial processes as well as other electrical machines. In recent years, the analysis of vibration signals has emerged as one of the most promising solutions for detecting faults in transformers. Nonetheless, it is not a straightforward process because of the nonstationary properties of the vibration signals and their high-level noise, as well as their different features when the transformer operates under different conditions. Based on the previously mentioned points, the motivation of this work is to contribute a methodology that can detect different severities of short-circuited turns (SCTs) in transformers in both transient and steady-state operating regimes using vibration signals. The proposed approach consists of a wavelet-based denoising stage, a short-time Fourier transform (STFT)-based analysis stage for the transient state, a Fourier transform (FT)-based analysis stage for the steady-state, the application of two fault indicators, i.e., the energy index and the total harmonic distortion index, and two neural networks for automatic diagnosis. To evaluate the effectiveness of the proposed methodology, a modified transformer is used to experimentally reproduce different levels of SCTs, i.e., 0-healthy, 5, 10, 15, 20, 25, and 30 SCTs, in a controlled way. The obtained results show that the proposed approach can detect the fault condition, starting from an initial stage for consolidation and a severe stage to accurately assess the fault severity, achieving accuracy values of 90%.

A Comparative Study on Denoising Algorithms for Footsteps Sounds as Biometric in Noisy Environments

Biometrics is the automated identification of a person based on distinctive characteristics, such as fingerprints, face, voice, or the sound of footsteps. This last characteristic has significant challenges considering the background noise present in any real-life application, where microphones would record footsteps sounds and different types of noise. For this reason, it is crucial to consider not only the capacity of classification algorithms for recognizing a person using foostetps sounds, but also at least one stage of denoising algorithms that can reduce the background sounds before the classification. In this paper we study the possibilities of a two-stage approach for this problem: a denoising stage followed by a classification process. The work focuses on discovering the proper strategy for applying combinations of both stages for specific noise types and levels. Results vary according to the type and level of noise, e.g., for White noise at signal-to-noise ratio level, accuracy can increase from 0.96 to 1.00 by applying deep learning based-filters, but the same option does not benefit the cases of signals with low level natural noises, where Wiener filtering can increase accuracy from 0.6 to 0.77 at the highest level of noise. The results represent a baseline for developing real-life implementations of footstep biometrics.

De-noising imaging through diffusers with autocorrelation.

Recovering targets through diffusers is an important topic as well as a general problem in optical imaging. The difficulty of recovering is increased due to the noise interference caused by an imperfect imaging environment. Existing approaches generally require a high-signal-to-noise-ratio (SNR) speckle pattern to recover the target, but still have limitations in de-noising or generalizability. Here, featuring information of high-SNR autocorrelation as a physical constraint, we propose a two-stage (de-noising and reconstructing) method to improve robustness based on data driving. Specifically, a two-stage convolutional neural network (CNN) called autocorrelation reconstruction (ACR) CNN is designed to de-noise and reconstruct targets from low-SNR speckle patterns. We experimentally demonstrate the robustness through various diffusers with different levels of noise, from simulative Gaussian noise to the detector and photon noise captured by the actual optical system. The de-noising stage improves the peak SNR from 20 to 38 dB in the system data, and the reconstructing stage, compared with the unconstrained method, successfully recovers targets hidden in unknown diffusers with the detector and photon noise. With the help of the physical constraint to optimize the learning process, our two-stage method is realized to improve generalizability and has potential in various fields such as imaging in low illumination.

An Efficient Multistage Approach for Blind Source Separation of Noisy Convolutive Speech Mixture

This paper proposes a novel efficient multistage algorithm to extract source speech signals from a noisy convolutive mixture. The proposed approach comprises two stages named Blind Source Separation (BSS) and de-noising. A hybrid source prior model separates the source signals from the noisy reverberant mixture in the BSS stage. Moreover, we model the low- and high-energy components by generalized multivariate Gaussian and super-Gaussian models, respectively. We use Minimum Mean Square Error (MMSE) to reduce noise in the noisy convolutive mixture signal in the de-noising stage. Furthermore, the two proposed models investigate the performance gain. In the first model, the speech signal is separated from the observed noisy convolutive mixture in the BSS stage, followed by suppression of noise in the estimated source signals in the de-noising module. In the second approach, the noise is reduced using the MMSE filtering technique in the received noisy convolutive mixture at the de-noising stage, followed by separation of source signals from the de-noised reverberant mixture at the BSS stage. We evaluate the performance of the proposed scheme in terms of signal-to-distortion ratio (SDR) with respect to other well-known multistage BSS methods. The results show the superior performance of the proposed algorithm over the other state-of-the-art methods.

Multinoise-type blind denoising using a single uniform deep convolutional neural network

Deep convolutional neural networks (CNNs) have achieved considerable success with image denoising. However, they still lack consistent performance across different noise types and levels. We extend noise scenarios to four categories: Gaussian, random-impulse, salt-and-pepper, and Poisson. We also propose a multinoise-type blind denoising network (MBDNet) that solves the blind denoising task using a uniform deep CNN architecture. The network can be divided into two stages where a concise CNN is first used to estimate auxiliary noise-type and noise-level information. Estimation results are then integrated as additional channels of the noisy image and are fed to the subsequent denoising stage. A unique two-branch structure is further adopted in the residual denoising CNN, wherein a shallow branch predicts the filter-flow mask and adaptively adjusts the feature extraction of the parallel deep branch. Extensive experiments on synthetic noisy images validate the effectiveness of the noise-estimation and denoising subnetworks and show that MBDNet is highly competitive as compared to state-of-the-art methods in both denoising performance and model runtime.

Noise Reduction for MEMS Gyroscope Signal: A Novel Method Combining ACMP with Adaptive Multiscale SG Filter Based on AMA.

In this paper, a novel hybrid method combining adaptive chirp mode pursuit (ACMP) with an adaptive multiscale Savitzky–Golay filter (AMSGF) based on adaptive moving average (AMA) is proposed for offline denoising micro-electromechanical system (MEMS) gyroscope signal. The denoising scheme includes preliminary denoising and further denoising. At the preliminary denoising stage, the original gyroscope signal is decomposed into signal modes one by one using ACMP with modified stopping criterion based on mutual information. Useful information is extracted while most noise is discarded in the residue at this stage. Then, AMSGF is proposed to further denoise the signal modes. Sample variance based on AMA is used to adjust the window size of AMSGF adaptively. Practical MEMS gyroscope signal denoising results under different motion conditions show the superior performance of the proposed method over empirical mode decomposition (EMD)-based denoising, discrete wavelet threshold denoising, and variational mode decomposition (VMD)-based denoising. Moreover, AMSGF is proven to gain a better denoising effect than some other common smoothing methods.

An efficient image denoising method based on principal component analysis with learned patch groups

We propose a novel principal component analysis (PCA)-based image denoising framework motivated by the fact that the PCA along with patch groups (PGs) can produce better denoising performance. The PGs essentially capture the geometric information from noisy image. In the denoising stage, the learned PGs are transformed to PCA domain and it only preserves the important principal components while removing the noise components. The proposed denoising method consists of mainly three stages such as patch grouping stage, dictionary learning stage and PCA-based denoising stage. Also, we present an algorithm Learned-PGPCA and tested it in a simulated environment. The experimental results divulged that the proposed denoising framework Learned-PGPCA achieved very competitive denoising performance, particularly in preserving edges and textures as compared to recent patch-based image denoising methods pertaining to gaussian noise.

Speech denoising using Bayesian NMF with online base update

A new speech denoising method based on online non-negative matrix factorization (NMF) is proposed in this paper. To achieve an efficient model for the temporal dependencies of speech and noise, and to improve the robustness for the actual non-stationary noisy environments, the Bayesian NMF is extended to the proposed model and a new noise basis matrix online update method is exploited. Firstly, the speech basis matrix is pre-trained off-line with the Bayesian NMF method. In speech denoising stage, the noise basis matrix is continuously updated by utilizing the noise frames in the noisy observation with the Bayesian NMF. The noise basis matrix is initialized via a pre-trained universal noise NMF model and the noise data for the matrix adaption are selected using a likelihood ratio test (LRT) speech decision criterion. Then the updated noise basis matrix and the pre-trained speech basis matrix are employed to the enhancement of the noisy signal. Finally, to address the incomplete separation and the speech distortion problem, a speech activity probability based noise suppression filter is presented to further eliminate the residue noise in the enhanced result. The experiment results show that the proposed method outperforms the comparative denoising algorithms in terms of objective measurement.

SIMPLE AND ROBUST EYE MOVEMENTS DETECTION METHOD

The aim of this research is to develop a signal processing method to detect four eye movements, such as looking up, down, left, right and blinking. This new method has a couple of features in comparison with the recent eye movement detection algorithms. Most of the recent algorithms require a De-noising stage, which is not required in this work. In addition, the suggested algorithm can be considered simple and robust in noisy environment in contrast with other algorithms. In this paper, short-time averaging method is proposed to process and to extract parameters from EOG signals. Moreover, adaptive threshold is applied to classify EOG pulses. The purpose of the adaptive threshold is to enhance the performance of the algorithm in a noisy background. Simulation results are based on real-life EOG signals, where these signals were recorded using an Electrooculography system. The results show that the proposed algorithm has a stable performance HR=100 % and FR=0% with SNR greater than 2dB. The average performance with SNR=0.5 dB is about HR= 90.21% and FR= 4.88%.

Sparse Representation Based Super-Resolution Algorithm using Wavelet Domain Interpolation and Nonlocal Means

In this correspondence, we propose a novel image resolution enhancement algorithm based on discretewavelet transform (DWT), stationary wavelet transform (SWT) and sparse signal recovery of the inputimage. The nonlocal means filter is employed in the preliminary denoising stage of the proposed method.The denoised input low resolution (LR) image is then decomposed into different frequency subbands byemploying DWT and SWT simultaneously. In parallel, the denoised LR image is subjected to a sparse signalrepresentation based interpolation. All the estimated high frequency subbands as well as the sparseinterpolated LR image are fused to generate a high resolution (HR) image by using inverse discrete wavelettransform (IDWT). Experimental results on various test images show the superiority of our method over theconventional and state-of-the-art single image super- resolution (SR) techniques in achieving the real timeperformance. Full Text: PDF DOI: http://dx.doi.org/10.11591/telkomnika.v16i2.8816

Sparse Representation Based Super-Resolution Algorithm using Wavelet Domain Interpolation and Nonlocal Means

In this correspondence, we propose a novel image resolution enhancement algorithm based on discretewavelet transform (DWT), stationary wavelet transform (SWT) and sparse signal recovery of the inputimage. The nonlocal means filter is employed in the preliminary denoising stage of the proposed method.The denoised input low resolution (LR) image is then decomposed into different frequency subbands byemploying DWT and SWT simultaneously. In parallel, the denoised LR image is subjected to a sparse signalrepresentation based interpolation. All the estimated high frequency subbands as well as the sparseinterpolated LR image are fused to generate a high resolution (HR) image by using inverse discrete wavelettransform (IDWT). Experimental results on various test images show the superiority of our method over theconventional and state-of-the-art single image super- resolution (SR) techniques in achieving the real timeperformance.

Fuzzy color video filtering technique for sequences corrupted by additive Gaussian noise

In this paper, a novel framework is presented for the denoising of color video sequences corrupted by additive Gaussian noise. The proposed technique consists of three filtering stages: spatial, spatio-temporal, and spatial postprocessing. During the first spatial stage, the gradient values in eight directions for pixels located in the vicinity of a central pixel, as well as the interchannel correlation between the analogous pixels in different color bands (RGB), are taken into account. These gradient values that estimate the level of noise contamination are employed using the designed fuzzy rules to preserve the image features (e.g., textures, edges, sharpness, and chromatic properties). In the spatio-temporal denoising stage, two consecutive video frames are filtered together, thereby yielding more information. Additionally, small local motions between consecutive frames are estimated using block matching procedure in different directions, gathering interframe samples with similar features for efficient denoising. In the final stage, the edge and plain areas in a current frame are separated for different spatial postprocessing denoising. Two variants of proposed fuzzy filter, depending on sliding windows, are proposed. Additionally, a hybrid fuzzy-Wiener denoising technique is performed employing the proposed filtering approach. Numerous simulation results confirm that these novel fuzzy frameworks outperform other state-of-the-art techniques in terms of objective criteria, as well as subjective visual perception in the various color sequences.

Peak Detection Using Wavelet Transform

A new work based-wavelet transform is designed to overcome one of the main drawbacks that found in the present new technologies. Orthogonal Frequency Division Multiplexing (OFDM)is proposed in the literature to enhance the multimedia resolution. However, the high peak power (PAPR) values will obstruct such achievements. Therefore, a new proposition is found in this work, making use of the wavelet transforms methods, and it is divided into three main stages; de-noising stage, thresholding stage and then the replacement stage. In order to check the system stages validity; a mathematical model has been built and its checked after using a MATLAB simulation. A simulated bit error rate (BER) achievement will be compared with our previously published work, where an enhancement from 8×10 -1 to be 5×10 -1 is achieved. Moreover, these results will be compared to the work found in the literature, where we have accomplished around 27% PAPR extra reduction. As a result, the BER performance has been improved for the same bandwidth occupancy. Moreover and due to the de-noise stage, the verification rate has been improved to reach 81%. This is in addition to the noise immunity enhancement.

Study on temperature error processing technique for fiber optic gyroscope

Fiber optic gyroscope (FOG) has been widely used as satellite and automobile attitude sensor for its high reliability and light weight. However, environment temperature variation results in errors, which will decrease the precision of FOG. The purpose of this paper is to present an error processing technique for FOG to reduce the influence of temperature variation based on the error analysis of FOG and a set of temperature experiment results. The technique contains two stages: one is de-noising stage, a novel de-noising algorithm is proposed, which integrated the advantages of lifting wavelet transform and adaptive forward linear prediction; the other is modeling stage, which based on Elman neural network. The simulation results show that the noise of FOG's output can be eliminated effectively by using the proposed de-noising algorithm, which is called LWT–FLP algorithm; the drift of FOG's output can be compensated effectively by Elman neural network model. The final processing results indicate that the errors caused by varied temperature have been reduced effectively by the proposed error processing technique.

Temperature variation effects on stochastic characteristics for low-cost MEMS-based inertial sensor error

We examine the effect of varying the temperature points on MEMS inertial sensors’ noise models using Allan variance and least-squares spectral analysis (LSSA). Allan variance is a method of representing root-mean-square random drift error as a function of averaging times. LSSA is an alternative to the classical Fourier methods and has been applied successfully by a number of researchers in the study of the noise characteristics of experimental series. Static data sets are collected at different temperature points using two MEMS-based IMUs, namely MotionPakII and Crossbow AHRS300CC. The performance of the two MEMS inertial sensors is predicted from the Allan variance estimation results at different temperature points and the LSSA is used to study the noise characteristics and define the sensors’ stochastic model parameters. It is shown that the stochastic characteristics of MEMS-based inertial sensors can be identified using Allan variance estimation and LSSA and the sensors’ stochastic model parameters are temperature dependent. Also, the Kaiser window FIR low-pass filter is used to investigate the effect of de-noising stage on the stochastic model. It is shown that the stochastic model is also dependent on the chosen cut-off frequency.