Sparse Decomposition Theory Research Articles

Fast sparse Bayesian learning-based seismic resolution enhancement

Seismic high resolution processing plays a crucial role in seismic interpretation and reservoir characterization. However, the low and high frequency of seismic signal are lost owing to the effects of earth filtering and diffraction. The missing of low-frequency increases the energy of wavelet side lobe, which not only enhances the interference with adjacent events, but also lead to side lobe artifacts. High frequency attenuation is more serious, which significantly reduces the seismic resolution. In addition, the current resolution enhancement methods are difficult to achieve desirable result in noise contaminated environment. A novel resolution enhancement method that extends high and low frequencies simultaneously and improves the signal-to-noise ratio (SNR) under the framework of sparse decomposition theory is proposed. Firstly, an over-complete dictionary is constructed using non-zero phase Ricker wavelet and then the effective seismic signal is extracted by sparse decomposition using the fast sparse Bayesian learning (FSBL) algorithm. Secondly, the effective frequency range of the seismic signal is determined through a series of atoms obtained by sparse decomposition. Then, a desired broadband spectrum can be identified depending on this range. By approximating the broadband spectrum of seismic data, the weak high-frequency information can be extended. Furthermore, a correction term is introduced in frequency domain to suppress the side lobes of atoms. The side lobes are well suppressed without distorting the main lobe, meanwhile, both the low and high frequencies are extended. Finally, the validity of this method is verified by synthetic and field data. The results show that this method has better performance than the ordinary deconvolution and spectral whitening methods in processing high resolution data.

Speech Enhancement Based on Constrained Low-rank Sparse Matrix Decomposition Integrated with Temporal Continuity Regularisation

Speech enhancement in strong noise condition is a challenging problem. Low-rank and sparse matrix decomposition (LSMD) theory has been applied to speech enhancement recently and good performance was obtained. Existing LSMD algorithms consider each frame as an individual observation. However, real-world speeches usually have a temporal structure, and their acoustic characteristics vary slowly as a function of time. In this paper, we propose a temporal continuity constrained low-rank sparse matrix decomposition (TCCLSMD) based speech enhancement method. In this method, speech separation is formulated as a TCCLSMD problem and temporal continuity constraints are imposed in the LSMD process. We develop an alternative optimisation algorithm for noisy spectrogram decomposition. By means of TCCLSMD, the recovery speech spectrogram is more consistent with the structure of the clean speech spectrogram, and it can lead to more stable and reasonable results than the existing LSMD algorithm. Experiments with various types of noises show the proposed algorithm can achieve a better performance than traditional speech enhancement algorithms, in terms of yielding less residual noise and lower speech distortion.

A new noise reduction method based on re-weighted group sparse decomposition and its application in gear fault feature detection

Aiming at the problem that gear vibration signals are susceptible to noise and the difficulty of extracting fault features, this paper proposes a new noise reduction method based on re-weighted group sparse decomposition (RWGSD). RWGSD introduces group sparse mode decomposition theory to protect the structural information of signal components in the frequency domain. On this basis, vital components are screened according to the time-domain characteristics of fault information, and the re-weighted enhancement is carried out. The fault characteristics are easy to identify in the final noise reduction result. In addition, RWGSD defines two new indicators, cyclic re-weighted kurtosis (CRWK) and re-weighted cyclic intensity (RWCI). CRWK can assess the intensity of periodic characteristic components and has some resistance to strong impact interference. RWCI can evaluate the magnitude of fault information, overcoming the limitations of traditional noise reduction techniques that screen out vital components based on energy size. Numerical simulation and real-world experiment results show that the proposed method has excellent performance in noise removal, increases the reliability of gear fault feature detection, and has certain practical values.

Harmonic Elimination and Magnetic Resonance Sounding Signal Extraction Based on Matching Pursuit Algorithm

Magnetic resonance sounding (MRS) is a non-invasive, direct, and quantitative geophysical method for detecting groundwater, and has been widely used in groundwater survey, water resource assessment, and disaster water source forecasting. However, the MRS signal is weak (nV level) and highly susceptible to environmental noise, such as random noise and power-line harmonics, resulting in reduced quality of received data. Achieving reliable extraction of MRS signals under strong noise is difficult. To solve this problem, we propose a matching pursuit algorithm based on sparse decomposition theory for data noise suppression and MRS signal extraction. In accordance with the characteristics of the signal and noise, an oscillating atomic library is constructed as a sparse dictionary to realize signal sparse decomposition. A two-step denoising strategy is proposed to reconstruct the power-line harmonics and then extract the MRS signal. We simulated synthetic data with different signal-to-noise ratios (SNRs), relaxation times, and Larmor frequencies. Our results show that the proposed algorithm can effectively remove power-line harmonics and reduce random noise. SNR is significantly improved by up to 35.6 dB after denoising. The effectiveness and superiority of the proposed algorithm are further verified by the measured data and through comparison with the singular spectrum analysis algorithm and harmonic modeling cancellation algorithm.

Privacy-Preserving Video Fall Detection via Chaotic Compressed Sensing and GAN-Based Feature Enhancement

Currently, falling is the primary cause of injury and death of the elderly due to accidents, which seriously threatens the health and life of the elderly. Considering the phenomenon of video privacy disclosure in recent years, this article proposes a computer vision fall detection method to meet the needs of video privacy protection. The innovation here lies in compressed sensing (CS) visual privacy protection and GAN-based feature enhancement. There are three main steps, the video frame visual privacy protection with chaotic pseudo-random CS mechanism, the foreground extraction with improved low rank sparse decomposition theory, and the temporal and spatial feature enhancement and fall detection with improved-ACGAN `architecture. The experimental results on three open fall datasets show that the method can not only effectively detect the fall behavior in video, but also has high accuracy. At the same time, the overall operation speed of the algorithm increases with the increase of the number of compression layers.

Research on the Segmentation of Biomarker for Chronic Central Serous Chorioretinopathy Based on Multimodal Fundus Image.

At present, laser surgery is one of the effective ways to treat the chronic central serous chorioretinopathy (CSCR), in which the location of the leakage area is of great importance. In order to alleviate the pressure on ophthalmologists to manually label the biomarkers as well as elevate the biomarker segmentation quality, a semiautomatic biomarker segmentation method is proposed in this paper, aiming to facilitate the accurate and rapid acquisition of biomarker location information. Firstly, the multimodal fundus images are introduced into the biomarker segmentation task, which can effectively weaken the interference of highlighted vessels in the angiography images to the location of biomarkers. Secondly, a semiautomatic localization technique is adopted to reduce the search range of biomarkers, thus enabling the improvement of segmentation efficiency. On the basis of the above, the low-rank and sparse decomposition (LRSD) theory is introduced to construct the baseline segmentation scheme for segmentation of the CSCR biomarkers. Moreover, a joint segmentation framework consisting of the above method and region growing (RG) method is further designed to improve the performance of the baseline scheme. On the one hand, the LRSD is applied to offer the initial location information of biomarkers for the RG method, so as to ensure that the RG method can capture effective biomarkers. On the other hand, the biomarkers obtained by RG are fused with those gained by LRSD to make up for the defect of undersegmentation of the baseline scheme. Finally, the quantitative and qualitative ablation experiments have been carried out to demonstrate that the joint segmentation framework performs well than the baseline scheme in most cases, especially in the sensitivity and F1-score indicators, which not only confirms the effectiveness of the framework in the CSCR biomarker segmentation scene but also implies its potential application value in CSCR laser surgery.

A sparse dictionary learning-based denoising method for underwater acoustic sensors

The complex and changeable marine environment will cause the signal detected by underwater acoustic sensors to be extremely weak, and traditional signal processing methods are usually complicated. To improve the detection efficiency and application scope of underwater acoustic sensors, we use sparse decomposition theory and dictionary learning algorithms to directly study the denoising of single hydrophone signals under different sea conditions. Then we use pulse compression technology to solve the problem of traditional methods that long signal causes construction difficulties of the dictionary. The method in this paper can randomly construct a discrete cosine transform (DCT) dictionary without a priori information. The noisy signal is trained and updated via orthogonal matching pursuit (OMP) and method of optimal directions (MOD), and the reconstruction of the signal is completed according to the updated dictionary and sparse coefficients. The results demonstrate that the method in this paper can be applied to wideband long pulse signals, and the signal-to-noise ratio (SNR) gain can reach about 20 dB under different sea conditions. By comparing with the OMP algorithm, the proposed method can reduce the number of atoms so that improving the system performance, reduce the algorithm complexity, and improve the operation efficiency.

Privacy-Preserving In-Home Fall Detection Using Visual Shielding Sensing and Private Information-Embedding

Falls are the main cause of accidental injuries, and even death among elderly people, especially those who live alone in their homes. The absence of a reliable fall detection system has long been a serious problem for home health monitoring. A video surveillance system can be used to monitor elderly people at home to detect falls, but the traditional implementation of such intelligent detection falls short of personal privacy-related considerations; additionally, many people do not want to be watched in their homes. To solve this problem, we propose a fall detection system with visual shielding that can ensure the safety of elderly people in their homes while preserving their personal privacy. Multilayer compressed sensing is first used to achieve visually shielded video frames. By combining low-rank sparse decomposition theory with the improved local binary pattern on the three orthogonal planes, the object features are extracted from the shielded video frames. Finally, to compensate for the information lost in the compressed video to a certain extent, a private information-embedded classification model is proposed to identify fall-related behavior. The experimental results on two public fall datasets show that the proposed method delivers impressive accuracy and a low error rate while effectively distinguishing between fall- and nonfall-related behaviors in videos.

Privacy-preserving video fall detection using visual shielding information

In recent years, fall detection, especially for the elderly living alone at home, is a challenge in the field of computer vision and pattern recognition. However, there is a concern of loss of privacy in intelligent visual surveillance. In order to solve the contradiction between security surveillance and privacy protection in vision-based fall detection methods, we propose a concept named visual shielding, which can be applied to eliminate visual information but not reduce monitoring function. In the preprocessing, the multilayer compressed sensing is used to compress video frames to achieve visual shielding effect. Then, object region is separated from shielded videos by the low-rank and sparse decomposition theory, based on which to extract motion trajectory features of the object via the dense trajectory algorithm. Finally, the fall detection issue is transformed into the sparse recognition problem of the signal. Experimental results on three public fall databases show that the specificity and sensitivity of the proposed method can be maintained at an ideal level, which has the dual advantages of high privacy protection and high recognition accuracy.

Weak ultrasonic signal detection in strong noise

In order to solve the problem of extracting ultrasonic signals from strong background noise, a novel method, which is termed APSO-SD algorithm and based on improved adaptive particle swarm optimization (APSO) and sparse decomposition (SD) theory, is proposed in this paper. This method can convert the ultrasonic signal denoising problem into optimizing the function on the infinite parameter set. First, based on the sparse decomposition theory and the structural characteristics of ultrasonic signal, the objective function of particle swarm optimization algorithm and the reconstruction algorithm of the denoised signal are constructed, so that particle swarm optimization and ultrasonic signal denoising can be combined. Second, in order to improve the robustness of the proposed approach, an APSO algorithm is proposed. What is more, because particle swarm optimization algorithm can be used to optimize in continuous parameter space, and according to the empirical characteristics of the ultrasonic signals used in practical engineering, a continuous super complete dictionary for matching ultrasonic signals is established. Since the super complete dictionary is continuous, there are an infinite number of atoms in the established dictionary. The redundancy of dictionaries is enhanced by the method in this paper. Based on the fact that the inner product of the optimal atom and the ultrasonic signal is one and the inner product of the noise and the optimal atom is zero in the established dictionary, the objective optimization function of APSO-SD algorithm is established. Finally, the optimal atom is determined based on the optimization result of the objective function. In this way, the denoising ultrasonic signal can be reconstructed by using the optimal atom according to the reconstruction algorithm. The processing results of simulated ultrasonic signals and measured ultrasonic signals show that the proposed method can effectively extract weak ultrasonic signals from strong background noise whose signal-to-noise ratio is lowest, as low as-4 dB. In addition, compared with the adaptive threshold based wavelet method, the proposed method in this paper shows the good denoising performance. In this paper, it is demonstrated that the problem of ultrasonic signal denoising can be transformed into the optimization of constraint functions. Furthermore, the ability of the proposed APSO-SD algorithm to accurately recover signals from noisy acoustic signals is better than that of the common wavelet method.

An Improved Fruit Fly Optimization Algorithm and Its Application in Heat Exchange Fouling Ultrasonic Detection

Inspired by the basic theory of Fruit Fly Optimization Algorithm, in this paper, cat mapping was added to the original algorithm, and the individual distribution and evolution mechanism of fruit fly population were improved in order to increase the search speed and accuracy. The flowchart of the improved algorithm was drawn to show its procedure. Using classical test functions, simulation optimization results show that the improved algorithm has faster and more reliable optimization ability. The algorithm was then combined with sparse decomposition theory and used in processing fouling detection ultrasonic signals to verify the validity and practicability of the improved algorithm.

Low-rank and sparse decomposition based shape model and probabilistic atlas for automatic pathological organ segmentation.

One major limiting factor that prevents the accurate delineation of human organs has been the presence of severe pathology and pathology affecting organ borders. Overcoming these limitations is exactly what we are concerned in this study. We propose an automatic method for accurate and robust pathological organ segmentation from CT images. The method is grounded in the active shape model (ASM) framework. It leverages techniques from low-rank and sparse decomposition (LRSD) theory to robustly recover a subspace from grossly corrupted data. We first present a population-specific LRSD-based shape prior model, called LRSD-SM, to handle non-Gaussian gross errors caused by weak and misleading appearance cues of large lesions, complex shape variations, and poor adaptation to the finer local details in a unified framework. For the shape model initialization, we introduce a method based on patient-specific LRSD-based probabilistic atlas (PA), called LRSD-PA, to deal with large errors in atlas-to-target registration and low likelihood of the target organ. Furthermore, to make our segmentation framework more efficient and robust against local minima, we develop a hierarchical ASM search strategy. Our method is tested on the SLIVER07 database for liver segmentation competition, and ranks 3rd in all the published state-of-the-art automatic methods. Our method is also evaluated on some pathological organs (pathological liver and right lung) from 95 clinical CT scans and its results are compared with the three closely related methods. The applicability of the proposed method to segmentation of the various pathological organs (including some highly severe cases) is demonstrated with good results on both quantitative and qualitative experimentation; our segmentation algorithm can delineate organ boundaries that reach a level of accuracy comparable with those of human raters.

Auto Regulated Data Provisioning Scheme with Adaptive Buffer Resilience Control on Federated Clouds

We investigate the channel state information (CSI) in multi-input multi-output (MIMO) cooperative networks that employ the amplify-and-forward transmission scheme. Least squares and expectation conditional maximization have been proposed in the system. However, neither of these two approaches takes advantage of channel sparsity, and they cause estimation performance loss. Unlike linear channel estimation methods, several compressed channel estimation methods are proposed in this study to exploit the sparsity of the MIMO cooperative channels based on the theory of compressed sensing. First, the channel estimation problem is formulated as a compressed sensing problem by using sparse decomposition theory. Second, the lower bound is derived for the estimation, and the MIMO relay channel is reconstructed via compressive sampling matching pursuit algorithms. Finally, based on this model, we propose a novel algorithm so called sparsity adaptive expectation maximization (SAEM) by using Kalman filter and expectation maximization algorithm so that it can exploit channel sparsity alternatively and also track the true support set of time-varying channel. Kalman filter is used to provide soft information of transmitted signals to the EM-based algorithm. Various numerical simulation results indicate that the proposed sparse channel estimation technique outperforms the previous estimation schemes.

A bearing fault diagnosis method based on sparse decomposition theory

The bearing fault information is often interfered or lost in the background noise after the vibration signal being transferred complicatedly, which will make it very difficult to extract fault features from the vibration signals. To avoid the problem in choosing and extracting the fault features in bearing fault diagnosing, a novelty fault diagnosis method based on sparse decomposition theory is proposed. Certain over-complete dictionaries are obtained by training, on which the bearing vibration signals corresponded to different states can be decomposed sparsely. The fault detection and state identification can be achieved based on the fact that the sparse representation errors of the signal on different dictionaries are different. The effects of the representation error threshold and the number of dictionary atoms used in signal decomposition to the fault diagnosis are analyzed. The effectiveness of the proposed method is validated with experimental bearing vibration signals.

Comparison of the acoustic performance and positioning accuracy of three kinds of planar partial discharge ultrasonic array sensors

The detection of partial discharge (PD) sources in electrical equipment is very important for safe and stable operation of power systems as it allows for the timely elimination of insulation faults. Cross, circular and square array sensors are commonly used in the detection. In order to determine which form is the best, this study compares the acoustic performance and positioning accuracy of the different arrays. Firstly, this study does research on the quantitative evaluation of the acoustic performance of the planar ultrasonic array sensors. Sparse signal decomposition theory is then applied to find three different directions of arrival (DOAs) of the signal from the PD source to the array sensor. Next, using two of the DOAs (azimuth and pitch), the location of the PD is determined as the centre of a sphere whose surface is tangential to the three DOAs in three different planes. The positioning accuracy of the sensors is then derived. Finally, simulation results show that the acoustic performance of the circular array is the best. It also produces the most accurate location results. Therefore, a circular array is the optimal form to use for the construction of a planar array sensor.

Sparse Decomposition Based Aero-engine’s Bearing Fault Diagnosis

Aero-engine's rolling bearings work in harsh environment such as high temperature, high speed and sharply varying load, thus resulting in complex failure modes and diagnostic inefficient. A novel fault feature extraction technique based on sparse decomposition theory, stagewise matching morphological component analysis(SMMCA) is proposed to decompose compound multicomponent signal into its building blocks which contain the fault feature information. The sparse decomposition method can decompose a signal in a redundant dictionary and obtain the sparse representation of the signal by an optimization algorithm. In the spirit of the sparse decomposition theory, the SMMCA firstly constructs specialized dictionaries where the distinct components of the original signal can be represented sparsely, and then the coefficients of each subcomponent in the tailed dictionary are obtained through the stagewise orthogonal matching pursuit(St OMP) algorithm, lastly, the subcomponent is reconstructed by virtue of the specialized dictionary and the corresponding coefficients. With the proposed method, the vibration signal of the aero engine fault bearing can be diagnosed precisely. The effectiveness of the proposed method is verified by both the simulation and the experiment.

Joint estimation of frequency and azimuth using acoustic vector sensor signals based on sparse decomposition theory

Acoustic vector sensor can obtain more information of sound field compared with scalar hydrophone. The sparse decomposition theory was put forward in the 90s of last century, and it provides a simple, flexible, and self-adaptive representation method of signal. Through sparse decomposition theory, it can essentially reduce the cost of signal processing and improve the compression efficiency. Space-time array manifold is constructed through signal analysis of single acoustic vector sensor (AVS). Based on sparse decomposition theory, the frequency and azimuth estimation algorithm is proposed, the frequencies and azimuths of multi-targets are estimated simultaneously by the joint estimation algorithm. Results using simulated data received from single acoustic vector sensor are illustrated. The accurate estimation of multi-targets’ frequencies, azimuths, and signal amplitudes can be obtained using the estimation algorithm we deduced. Then, the influence of targets number, signal-to-noise (SNR), snapshots number on algorithm performance is analyzed.

Sparse Channel Estimation for MIMO-OFDM Two-Way Relay Network with Compressed Sensing

Accurate channel impulse response (CIR) is required for equalization and can help improve communication service quality in next-generation wireless communication systems. An example of an advanced system is amplify-and-forward multiple-input multiple-output two-way relay network, which is modulated by orthogonal frequency-division multiplexing. Linear channel estimation methods, for example, least squares and expectation conditional maximization, have been proposed previously for the system. However, these methods do not take advantage of channel sparsity, and they decrease estimation performance. We propose a sparse channel estimation scheme, which is different from linear methods, at end users under the relay channel to enable us to exploit sparsity. First, we formulate the sparse channel estimation problem as a compressed sensing problem by using sparse decomposition theory. Second, the CIR is reconstructed by CoSaMP and OMP algorithms. Finally, computer simulations are conducted to confirm the superiority of the proposed methods over traditional linear channel estimation methods.

Improved Sparse Channel Estimation for Cooperative Communication Systems

Accurate channel state information (CSI) is necessary at receiver for coherent detection in amplify-and-forward (AF) cooperative communication systems. To estimate the channel, traditional methods, that is, least squares (LS) and least absolute shrinkage and selection operator (LASSO), are based on assumptions of either dense channel or global sparse channel. However, LS-based linear method neglects the inherent sparse structure information while LASSO-based sparse channel method cannot take full advantage of the prior information. Based on the partial sparse assumption of the cooperative channel model, we propose an improved channel estimation method with partial sparse constraint. At first, by using sparse decomposition theory, channel estimation is formulated as a compressive sensing problem. Secondly, the cooperative channel is reconstructed by LASSO with partial sparse constraint. Finally, numerical simulations are carried out to confirm the superiority of proposed methods over global sparse channel estimation methods.