Noise Reduction Processing Research Articles

Case study of hydraulic fracture monitoring using multiwell integrated analysis based on low-frequency DAS data

Distributed acoustic sensing (DAS) is an effective technique for hydraulic fracture monitoring. It can potentially constrain fracture propagation direction and time while monitoring strain perturbation, such as stress shadowing. In this study, we acquired passive DAS and distributed temperature sensing (DTS) data throughout the entire fracturing operations of adjacent production wells with varying offset lengths from the fiber-optic cable in the Montney tight gas area. We applied data processing techniques to the DAS data to extract low-frequency components (less than 0.5 Hz) and to construct the strain rate and cumulative strain maps for detecting responses related to fracture hits along the fiber-optic cable. We used low-frequency DAS (LF-DAS) results to estimate the fracture hit position and time, and in certain cases, to additionally estimate the fracture connection. By integrating LF-DAS results with DTS results, we detected the temperature changes around the compression response near the fracture hit position and time. Furthermore, we observed that timing of the fracture hit can be constrained more precisely by using high-frequency DAS data (greater than 10 Hz). We estimated the fracture propagation direction and speed from the estimated fracture hit position and time. The fracture propagation direction deviated slightly from a perpendicular line to the fiber direction. In addition, as estimated from the first fracture hit time, the fracture length and fluid injection volume had a proportional relationship. Due to challenges associated with the data, it is important to design data acquisition geometry and fracturing operations on the premise of acquiring LF-DAS data. It is also important to apply an additional noise reduction process to the data.

Supervised Retinal Vessel Segmentation Based Average Filter and Iterative Self Organizing Data Analysis Technique

Retinal fundus is the inner surface of the eye associated with the lens. The identification of disease needs some parts of retinal fundus, such as blood vessel. Blood vessels are part of circulation system which functions to supply blood to retina area. This research proposed a method for segmentation of blood vessel in retinal image with Average Filter and Iterative Self-Organizing Data Analysis (ISODATA) Technique. The first step with the input image changed to Gamma Correction, increasing contrast with Contrast Limited Adaptive Histogram Equalization (CLAHE), the filtering process with Average Filter. The segmentation is used for ISODATA. Region of Interest was applied to take the center of a vessel object and remove the background. In the final stage, the process of noise reduction and removal of small pixel values with Median Filter and Closing Morphology. Datasets used in this research were DRIVE and STARE. The average result was obtained for STARE dataset with an accuracy of 94.41%, Sensitivity of 55.57%, Specification of 98.31%, F1 Score of 64.81% while for the DRIVE dataset with accuracy of 94.78%, Sensitivity of 43.46%, Specification of 99.81%, and F1 Score of 59.39%.

Research on a Noise Reduction Method Based on DTCWT and the Cyclic Singular Energy Difference Spectrum

The gear is the most important part of the transmission system of mechanical equipment, and the monitoring and diagnosis of it can improve the reliability of mechanical equipment. However, mechanical equipment generally works in harsh working conditions. The gear vibration signal is subjected to strong noise interference in working conditions, which brings great challenges for the effective diagnosis of gear faults. This paper proposed a noise reduction method based on the dual-tree complex wavelet transform (DTCWT) and cyclic singular energy difference spectrum. First, the gear vibration signal containing strong noise interference is decomposed into a series of signal components with different frequency characteristics by using the time-frequency analysis ability of DTCWT. Then, cyclic singular energy difference spectrum is proposed based on the idea of a cascaded cycle and the successive elimination of noise interference to process each signal component with different frequency characteristics, and the termination conditions of cyclic singular energy difference spectrum can be set according to the noise interference distribution characteristics in different frequency bands. The final noise reduction of the original gear vibration signal can be realized based on signal reconstruction after the noise reduction processing of each signal components with different frequency bands. Finally, experiments are carried out to verify the effectiveness of the proposed method, which is effective and suitable for the noise reduction of the vibration signal.

Combining GNSS and accelerometer measurements for evaluation of dynamic and semi-static characteristics of bridge structures

With the increasing number of long span bridges, real-time, accurate and continuous monitoring of their safety is important at present. This study investigates the combination of a global navigation satellite system (GNSS) and accelerometer for monitoring dynamic and semi-static characteristics of bridge structures. A field experiment was conducted with the integration of a GNSS and accelerometer. Considering the noise interference of GNSS monitoring, performance tests were first conducted in different environments to investigate the noise characteristics. Next, complete ensemble empirical mode decomposition with adaptive noise-wavelet packet (CEEMDAN-WP) algorithm was chosen for denoising, among which a double criterion based on the correlation coefficient and effective coefficient was proposed to sift the intrinsic mode functions. After the noise reduction process, structural dynamic displacements and modal frequencies were successfully extracted from the 50 Hz GNSS real-time kinematic (GNSS-RTK) and accelerometer data, in which the displacements presented a consistent trend and the first natural frequency was the same (i.e. 0.369 Hz). Structural semi-static characteristics were evaluated by using 1 Hz (RTK), post-processed kinematic, and precise point positioning data. With reference to relevant specifications, the structural failure probability of the bridge in the vertical direction was calculated to be 0.4319. The results indicate that GNSS-RTK is reliable in monitoring structural dynamic and semi-static displacements of the bridge. Additionally, the proposed improved CEEMDAN-WP with double criterion is effective for background noise reduction. In addition, there may be some non-adequate behaviors, such as heavy traffic and vehicle overload, leading to the critical operation of the bridge.

Using a dual-tree complex wavelet transform for denoising an optical coherence tomography angiography blood vessel image

High image quality is of great importance for precise diagnosis and therapeutics of eye disease in clinic. A human retina OCT angiography (OCTA) image can be extracted from multiple OCT B-scans to visualize the distribution of blood vessels. However, OCTA suffer from the degeneration of image quality due to inherent Gaussian noise of the OCT system while the blood vessel’s signal is extracted. The degeneration of the noise in OCTA image will be more conducive to the evaluation of abnormal and normal blood vessels in the human eye. To precisely assist diagnosis and therapeutics in clinic by reducing the Gaussian noise in the OCTA image, an OCTA image denoising method is proposed based on the dual-tree complex wavelet transform and bilateral shrinking Bayes frame. Initially, OCTA images are extracted from the raw data based on the optical microangiography algorithm. Then, the image is decomposed into the wavelet domain using the dual-tree complex wavelet transform. The signal and noise among different wavelet scale layers are separated on the basis of the Bayesian posterior probability. Finally, the inverse wavelet transform is employed to reconstruct the denoised image. Through the noise reduction process of the algorithm, the PSNR and CNR of the OCTA image are increased by 49.15% and 47.91%, respectively. According to the results, the wavelet transform can effectively separate the blood flow signal and noise in processing the OCTA signal, which will provide an effective image processing method for the clinical evaluation requiring high-quality OCTA images.

Application of Pseudo-color Image Feature-Level Fusion in Nondestructive Testing of Wire Ropes

Among the many nondestructive testing methods of ferromagnetic materials, magnetic flux leakage (MFL) testing is a very mature technology. Due to the single source of target information in traditional MFL detection technology, the recognition accuracy is low. Infrared image is a color artificially given, so it is also a pseudo-color image. We use the feature-level image fusion technology to fuse color infrared images and pseudo-color MFL images and quantitatively identify the defect feature vectors after fusion. A wavelet noise reduction algorithm is used to perform noise reduction processing on the MFL signal and used the pseudo-color transform method to convert the noise-reduced MFL signal into a pseudo-color image. For the collected infrared images of the wire rope, we use an image segmentation algorithm to extract the defective parts. The color moments and texture features of the pseudo-color MFL defect image and the color infrared defect image are extracted for feature-level fusion and trained as the input of the nearest neighbor algorithm. The final experimental results prove that the fusion of the two images information has achieved good results and the defect recognition rate has been improved.

Noise Reduction Model of Blasting Seismic Wave Signal

The existence of noise will make the results of time-frequency analysis of blasting seismic wave lack of authenticity. In order to obtain the real blasting vibration characteristics, a method based on the complementary ensemble empirical mode decomposition with adaptive noise and establishing objective function is proposed. The process of noise reduction is divided into three steps. Firstly, the noise signal is decomposed by the complete ensemble empirical mode decomposition with adaptive noise, and the noise reduction algorithm is established based on the intrinsic mode function obtained from the decomposition; secondly, the objective function considering the smoothness of the noise reduction algorithm and its correlation with the measured signal is established; finally, the algorithm corresponding to the optimal solution of the objective function is found, which is the noise reduction model of blasting seismic wave signal. The model is applied to the de-noising of blasting seismic signal, and the de-noising ability of the model is analyzed by the noise reduction error ratio. The analysis results show that the model can reduce the noise of the seismic monitoring signal with noise on the premise of fully preserving the real components of the blasting seismic signal.

Seismic exploration desert noise suppression based on complete ensemble empirical mode decomposition with adaptive noise

A foundation for the imaging and interpretation of stratigraphic structures is the high quality seismic data. The desert seismic events are contaminated by strong random noise. The random noise in desert seismic records has complex features, including nonlinear, non-Gaussian and low frequency, which is different from the random noise in people's previous cognition. In addition, the effective signal of desert seismic record and noise has the same frequency band. These situations have brought great difficulties to the denoising of conventional methods. While the conventional denoising algorithm for dealing with seismic signals is mainly for the suppression of random Gaussian noise. So the suppression result of random desert noise is not ideal. Based on these problems, we introduce the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) algorithm into the noise reduction process of desert seismic records. Compared with other time-frequency analysis methods, CEEMDAN algorithm has significant feature recognition effect, and remains the characteristics of adaptability. At the same time, it overcomes the problems of modal aliasing and false components. The experimental results show that denoising effect is obviously better than those of conventional denoising methods, and the suppression of the surface wave in the real seismic record is relatively thorough.

Impulse noise reduction using hybrid neuro‐fuzzy filter with improved firefly algorithm from X‐ray bio‐images

AbstractNoise filtering performance in medical images is improved using a neuro‐fuzy network developed with the combination of a post processor and two neuro‐fuzzy (NF) filters. By the fact, the Sugeno‐type is found to be less accurate during impulse noise reduction process. In this paper, we propose an improved firefly algorithm based hybrid neuro‐fuzzy filter in both the NF filters to improve noise reduction performance. The proposed noise reduction system combines the advantages of the neural, fuzzy and firefly algorithms. In addition, an improved version of firefly algorithm called searching diversity based particle swarm firefly algorithm is used to reduce the local trapping problem as well as to determine the optimal shape of membership function in fuzzy system. Experimental results show that the proposed filter has proved its effectiveness on reducing the impulse noise in medical images against different impulse noise density levels.

Dance Movement Interference Suppression Algorithm Based on Wearable Sensors in a Smart Environment

<p style="text-align: justify;"><span style="font-size: 16px; font-family: "times new roman";">Wearable device (Wearable device) is a hardware device that can be installed on clothes, clothing accessories or directly worn on the body for easy carrying. It can collect a large amount of user data and behavior habits in real time, and complete some by using software information technology. This paper aims to study the dance movement interference suppression algorithm based on wearable sensors in a smart environment. This paper proposes a method of human action recognition based on wearable sensors, using the original motion capture data collected by the wearable sensors to perform three-dimensional reconstruction, which shows that the basic link of constructing action words is the extraction of key gestures. In addition, this article also proposed an interference suppression algorithm, introduced wavelet noise reduction processing, calculated wavelet coefficients, noisy signals and other basic information, which proved the effectiveness of the interference suppression algorithm. The experimental results of this article show that the wearable sensor dance movement recognition in a smart environment has been greatly improved compared with the traditional dance movement recognition. Among them, the recognition rate of dance movements has increased by 25%, and the error of dance movements has become smaller and smaller. The results show that the interference suppression algorithm has obvious effects on the dance movements of the wearable sensor.</span>

A new joint noise reduction and echo suppression system based on FBSS and automatic voice activity detector

In hands-free communication framework, the main drawback affecting the system is the presence of the acoustic noise and echo interferences. To address these problems at the same time, we propose a new joint noise reduction and echo suppression system based on forward blind source separation (FBSS) structure and automatic voice activity detection (AVAD). To deal with the acoustic echo issue, we propose to use an echo suppressor (ES) based on the single channel normalized least mean square (SCNLMS) algorithm. In order to obtain an accurate estimate of the echo signal, the punctual noise is pre-processed by a noise reduction (NR) process, where, for the NR process, we propose to use the two channel normalized least mean square (TCNLMS) algorithm based on the FBSS structure. The FBSS structure uses often manual voice activity detection (MVAD) mechanism to separate speech from noise, but this is not practice in real situations. To overcome the use of the MVAD, we propose a system that uses a new automatic voice activity detection (automatic VAD) mechanism based on signal-to-noise-ratio (SNR) estimation. The obtained results in terms of objective criteria have shown the efficiency of the proposed system in dealing with both interferences at the same time.

Fault Recovery Path Analysis of a Software Dynamic Image Based on a Fuzzy Control Algorithm

In order to improve the ability of software dynamic image fault detection, a software dynamic image fault recovery path detection algorithm based on a fuzzy control algorithm is proposed. A software dynamic image fault signal model of a software dynamic image fault was constructed by adopting an embedded feature extraction and a fuzzy control algorithm, and the dynamic image fault signal of the embedded software under the multi-load was subjected to frequency spectrum decomposition and blind source separation. The method comprises the following steps: (1) carrying out noise reduction processing on a software dynamic image fault signal by adopting a multi-dimensional wavelet decomposition method, (2) carrying out wavelet entropy feature extraction on the software dynamic image fault signal of the noise reduction output, and (3) combining the wavelet structural feature recombination method to carry out the recombination of the software dynamic image fault feature. The high-order spectral characteristic of the software dynamic image fault signal was extracted and the high-order spectral characteristic of the extracted embedded software dynamic image fault recovery path was automatically matched, and the automatic identification and detection of the fault part of the software dynamic image was realized.

An Air Quality Prediction Model Based on a Noise Reduction Self-Coding Deep Network

Aiming at remedying the problem of low prediction accuracy of existing air pollutant prediction models, a denoising autoencoder deep network (DAEDN) model that is based on long short-term memory (LSTM) networks was designed. This model created a noise reduction autoencoder with an LSTM network to extract the inherent air quality characteristics of original monitoring data and to implement noise reduction processing on monitoring data to improve the accuracy of air quality predictions. The LSTM network structure in the DAEDN model was designed as bidirectional LSTM (Bi-LSTM) to solve the problem of a lag in the unidirectional LSTM prediction results and thereby to further improve the prediction accuracy of the prediction model. Using air pollutant time series data, the DAEDN model was trained using hourly PM2.5 concentration data collected in Beijing over 5 years. The experimental results show that the DAEDN model can extract more stable features from the noisy input after training was completed. The models were evaluated using RMSE and MAE, and the results show that the indexes are 15.504 and 6.789; compared with unidirectional LSTM, it is reduced by 7.33% and 5.87%, respectively. In addition, the new prediction model essentially considered the time series properties of the prediction of the concentration of spatial pollutants and the fully integrated environmental big data, such as air quality monitoring, meteorological monitoring, and forecasting.

Development of a mathematical model and analysis of the receiver for emergency alerts in a mine

The article presents the scientific and methodological approaches to the principles of construction and justification of the characteristics of a new sample of the SDV radio channel in underground mines to assess the required sensitivity of the receiving equipment and the maximum range of the channel. Based on the estimates obtained, as well as using radio engineering methods for calculating communication systems, the principles of noise reduction are formulated, coding and modulation systems are selected and justified. Mathematical modeling of noise reduction processes was carried out and algorithms and programs for the receiving and transmitting paths of a wireless data transmission system through rocks were developed. The paper presents the results of field observations of noise recorded in a working mine for the extraction of non-ferrous metals.

Using Hilbert-Huang Transform to Process and Analyze the Corrosion Acoustic Emission Signal of the Tank Bottom Plate

In this article, Hilbert-Huang transform (HHT) is applied to the tank bottom corrosion acoustic emission signal de-noising processing and decomposition, the effective signals are explored and extracted, and a new method of noise reduction for acoustic emission signal is proposed. The acoustic emission signal of corrosion of tank bottom plate is processed and decomposed by Hilbert-Huang transform, and the energy ratio of each sub-signal in the original signal is calculated, the source data of each sub-signal is extracted, and the effective acoustic emission signal in the original signal is extracted. Using the calculation formula of acoustic emission amplitude, the amplitude of each sub-signal is calculated respectively. The results show that the amplitude of each sub-signal is smaller than that of the original signal. This method can be used to reduce noise and extract effective acoustic emission signals. And it provides the foundation support, to form the method of noise reduction processing and effective signal extraction for corrosion acoustic emission signal on the tank bottom plate. It has some application value to improve the accuracy of the evaluation results of the corrosion state of the tank bottom plate.

Acoustic emission propagation characteristics and damage source localization of asphalt mixtures

Fracture of asphalt materials has been a critical issue that affects the fatigue performance and service life of asphalt pavements. Acoustic emission (AE) technique, as a kind of non-destructive testing (NDT) method, can effectively detect minor damage in various materials. The objective of this paper was to propose a damage source localization method suitable for understanding the fracture behavior of asphalt mixtures by AE detection. Firstly, the pencil-lead-break (PLB) tests were conducted on asphalt mixture beams to explore the propagation characteristics of AE signals, and a reasonable layout scheme of AE sensors was determined for monitoring the AE process of asphalt mixtures. Secondly, the wavelet threshold denoising method was utilized to extract effective information from AE signals associated with the pencil-lead breaks on the surface of asphalt mixture beam by simultaneously considering the wavelet basis functions, decomposition levels and threshold rules. Finally, an effective time difference of arrival (TDOA) estimation method combining noise reduction, wavelet decomposition and cross-correlation processing was established to accurately locate the damage source of asphalt mixtures. The results show that there is less serious attenuation of amplitude and energy of AE parameters and frequency spectrum of AE waveforms within the range of 100 mm in asphalt mixtures. The larger signal-to-noise ratio (SNR) and the smaller root mean square error (RMSE) of denoised AE signals indicate that the threshold denoising method optimized by the Fruit Fly Optimization Algorithm (FOA) is more effective for AE detection of asphalt mixtures. The improved TDOA estimation method could obtain a minimum TDOA value, and the calculated locations of AE events are closer to the actual PLB points.

Research on sports video detection technology motion 3D reconstruction based on hidden Markov model

The difficulty of sports video detection technology lies in how to detect the end point segment from the complex video speech environment, and the artificial intelligence technology is still in the research stage. Based on this, this study builds a model based on the hidden Markov model. At the same time, the video file noise reduction processing is performed by the spectral subtraction noise reduction algorithm of the complex domain extension. Moreover, combined with the actual situation of sports competitions, this paper proposes an endpoint detection algorithm based on variance characteristics, and comprehensively designs a speech recognition model based on Markov model. In order to verify the validity of the model, the performance of the model is verified by an example, and the real sports competition is taken as the research object, and the accuracy rate and the recall rate are set as performance indicators. The research shows that the model proposed in this study performs well in both accuracy and performance rate and can be used as a reference for artificial intelligence application to sports video detection technology.

Determining Philippine coconut maturity level using machine learning algorithms based on acoustic signal

Advanced intelligent systems are becoming significant to many sectors, including farming. In agriculture, the intelligent classification of post-harvested fruits seems to have a direct impact on farmers, mainly for export products. Unlike other popular fruits, coconuts tend to have limited studies due to its tropical nature grown in developing countries as well as its unique physical structure. In this study, a classification of real coconut datasets is performed based on acoustic signals acquired through a developed tapping system and learned by three widely-used machine learning techniques - artificial neural network (ANN), random forest (RF) and support vector machine (SVM). There are 129 coconuts samples, each classified into three maturity levels – pre-mature, mature, and over-mature. A three-second tapping system gathered from each sample a total of 132,300 data points, which underwent noise reduction and signal processing. Each machine learning model predicts the class of the fruit by learning the patterns of the transformed frequency spectrums of each sample signal. Based on ten times cross-validated results, the three machine learning algorithms satisfactorily predicted the maturity level of coconuts with at least 80% classification accuracy. All models correctly predicted over-mature coconuts but confused in classifying pre-mature with mature and mature with over-mature coconuts. RF model outperformed the other models with efficiencies of 90.98% and 83.48% accuracies for training and testing, respectively. The imbalance data for each coconut class can be addressed to give better results. Additionally, the prepared coconut dataset may use more advanced deep learning techniques.

Using a Machine Learning Algorithm Integrated with Data De-Noising Techniques to Optimize the Multipoint Sensor Network.

In this paper, for an intensity wavelength division multiplexing (IWDM)-based multipoint fiber Bragg grating (FBG) sensor network, an effective strain sensing signal measurement method, called a long short-term memory (LSTM) machine learning algorithm, integrated with data de-noising techniques is proposed. These are considered extremely accurate for the prediction of very complex problems. Four ports of an optical coupler with distinct output power ratios of 70%, 60%, 40%, and 30% have been used in the proposed distributed IWDM-based FBG sensor network to connect a number of FBG sensors for strain sensing. In an IWDM-based FBG sensor network, distinct power ratios of coupler ports can contain distinct powers or intensities. However, unstable output power in the sensor system due to random noise, harsh environments, aging of the equipment, or other environmental factors can introduce fluctuations and noise to the spectra of the FBGs, which makes it hard to distinguish the sensing signals of FBGs from the noise signals. As a result, noise reduction and signal processing methods play a significant role in enhancing the capability of strain sensing. Thus, to reduce the noise, to improve the signal-to-noise ratio, and to accurately measure the sensing signal of FBGs, we proposed a long short-term memory (LSTM) deep learning algorithm integrated with discrete waveform transform (DWT) data smoother (de-noising) techniques. The DWT data de-noising methods are important techniques for analyzing and de-noising the sensor signals, and it further improves the strain sensing signal measurement accuracy of the LSTM model. Thus, after de-noising the sensor data, these data are fed into the LSTM model to measure the sensing signal of each FBG. The experimental results prove that the integration of LSTM with the DWT data de-noising technique achieved better sensing signal measurement accuracy, even in noisy data or environments. Therefore, the proposed IWDM-based FBG sensor network can accurately sense the signal of strain, even in bad or noisy environments; can increase the number of FBG sensors multiplexed in the sensor system; and can enhance the capacity of the sensor system.

Particle Swarm Optimization Algorithm with Mutation Operator for Particle Filter Noise Reduction in Mechanical Fault Diagnosis

In this paper, a new particle swarm optimization particle filter (NPSO-PF) algorithm is proposed, which is called particle cluster optimization particle filter algorithm with mutation operator, and is used for real-time filtering and noise reduction of nonlinear vibration signals. Because of its introduction of mutation operator, this algorithm overcomes the problem where by particle swarm optimization (PSO) algorithm easily falls into local optimal value, with a low calculation accuracy. At the same time, the distribution and diversity of particles in the sampling process are improved through the mutation operation. The defect of particle filter (PF) algorithm where the particles are poor and the utilization rate is not high is also solved. The mutation control function makes the particle set optimization process happen in the early and late stages, and improves the convergence speed of the particle set, which greatly reduces the running time of the whole algorithm. Simulation experiments show that compared with PF and PSO-PF algorithms, the proposed NPSO-PF algorithm has lower root mean square error, shorter running time, higher signal-to-noise ratio and more stable filtering performance. It is proved that the algorithm is suitable for real-time filtering and noise reduction processing of nonlinear signals.