Independent Component Analysis Processing Research Articles

Analysis of the Spatial Distribution and Common Mode Error Correlation in a Small-Scale GNSS Network.

When analyzing GPS time series, common mode errors (CME) often obscure the actual crustal movement signals, leading to deviations in the velocity estimates of station coordinates. Therefore, mitigating the impact of CME on station positioning accuracy is crucial to ensuring the precision and reliability of GNSS time series. The current approach to separating CME mainly uses signal filtering methods to decompose the residuals of the observation network into multiple signals, from which the signals corresponding to CME are identified and separated. However, this method overlooks the spatial correlation of the stations. In this paper, we improved the Independent Component Analysis (ICA) method by introducing correlation coefficients as weighting factors, allowing for more accurate emphasis or attenuation of the contributions of the GNSS network's spatial distribution during the ICA process. The results show that the improved Weighted Independent Component Analysis (WICA) method can reduce the root mean square (RMS) of the coordinate time series by an average of 27.96%, 15.23%, and 28.33% in the E, N, and U components, respectively. Compared to the ICA method, considering the spatial distribution correlation of stations, the improved WICA method shows enhancements of 12.53%, 3.70%, and 8.97% in the E, N, and U directions, respectively. This demonstrates the effectiveness of the WICA method in separating CMEs and provides a new algorithmic approach for CME separation methods.

Noise source separation and identification of excavator cab based on EEMD-PCA-ICA-ST

In order to effectively separate and identify the noise source of excavator cab, this paper adopts the methods of ensemble empirical mode decomposition (EEMD), principal component analysis (PCA), independent component analysis (ICA) and S-transform (ST) to study. Firstly, EEMD is used to decompose the single channel noise signal collected at the driver’s position into multiple intrinsic mode functions (IMF); Secondly, PCA is used to reduce the dimension of IMFs to remove the redundant IMF; Next, ICA processing is performed on the remaining IMFs to obtain independent IC components; Finally, S-transform is used to analyze the time-frequency of IC component, and the source of IC component is identified by prior experience. The results show that the main noise sources are the vibration radiation noise of related parts in the cab, the combustion noise of diesel engine and the fan noise. After identifying the main noise sources, the next step is to study the contribution of these noise sources, which provides a new idea for noise control.

3Pb6-5 Relationship between target position and processing waveform related to ICA (Independent Component Analysis) processing

3Pb6-5 Relationship between target position and processing waveform related to ICA (Independent Component Analysis) processing

NIRS-ICA: A MATLAB Toolbox for Independent Component Analysis Applied in fNIRS Studies.

Independent component analysis (ICA) is a multivariate approach that has been widely used in analyzing brain imaging data. In the field of functional near-infrared spectroscopy (fNIRS), its promising effectiveness has been shown in both removing noise and extracting neuronal activity-related sources. The application of ICA remains challenging due to its complexity in usage, and an easy-to-use toolbox dedicated to ICA processing is still lacking in the fNIRS community. In this study, we propose NIRS-ICA, an open-source MATLAB toolbox to ease the difficulty of ICA application for fNIRS studies. NIRS-ICA incorporates commonly used ICA algorithms for source separation, user-friendly GUI, and quantitative evaluation metrics assisting source selection, which facilitate both removing noise and extracting neuronal activity-related sources. The options used in the processing can also be reported easily, which promotes using ICA in a more reproducible way. The proposed toolbox is validated and demonstrated based on both simulative and real fNIRS datasets. We expect the release of the toolbox will extent the application for ICA in the fNIRS community.

Improve the performance of independent component analysis by mapping the spectrum to an orthogonal space

Independent Component Analysis (ICA) has attracted chemists recently, for its charm can separate the independent signals from a mixed system and does not need prior knowledge. However, its dissatisfactory performance for the chemical measured signal is still blocking the practicability. Thus, this paper summarized the ICA processing path from the establishment of rectangular coordinates in linear space to the determination of the corresponding relation between the coordinate system and real components. The primary cause of the deviation between the ICA results and the chemical measurements is that the measuring signal was subject to uncertainty. Besides, uncertainty made the deviation of source signal from the statistical independence assumption, or in other words, it appeared to be nonorthogonal. For this key, it proposed to map the measured value to the high-order derivative space, use the derivative to narrow the peak width, reduce the influence of uncertainty, and improve the separation performance of ICA to chemical measurement signal, such as the spectrum. Actual cases of this paper showed that when up to 6th order, the separating results had been perfect for IR spectra, and even for homologs isomers.

Robust data processing of noisy marine controlled-source electromagnetic data using independent component analysis

Data processing techniques are often used to estimate the noise-free response of marine controlled-source electromagnetic (CSEM) data and magnetotelluric transfer functions. We have implemented a new CSEM data processing scheme that uses a robust method based on independent component analysis (ICA) to extract interpretable datasets from noisy marine CSEM data. We applied the data processing scheme to signals from a new CSEM observation system comprising a remotely operated vehicle (ROV) and an ocean bottom electromagnetometer (OBEM). These datasets were obtained around the Iheya North hydrothermal field, Okinawa Trough, Japan. The observation system allows a small-scale CSEM survey to be conducted in areas of steep topography, such as hydrothermal fields, because the ROV can deploy the OBEM at the exact observation site. The results show that the coherent and environment noise that exists in the raw time series is reduced sufficiently by ICA processing. It makes interpretation of the resulting electric field data possible. The results also show that the processed data has a higher signal-to-noise ratio in the middle-to-high-frequency band than the data without ICA. The normalised spectrum, obtained by normalising the observed data from the hydrothermal area, indicates that a conductive anomaly exists in the near-offset area around the OBEM. We apply 2D inversion to the electric field data and find that a low resistivity body exists beneath the OBEM and 50 m offset from the OBEM. This resistivity structure is consistent with images taken by the ROV that show characteristic organisms in hydrothermal seepage around the OBEM site.

High- and low-level hierarchical classification algorithm based on source separation process

High-dimensional data applications have earned great attention in recent years. We focus on remote sensing data analysis on high-dimensional space like hyperspectral data. From a methodological viewpoint, remote sensing data analysis is not a trivial task. Its complexity is caused by many factors, such as large spectral or spatial variability as well as the curse of dimensionality. The latter describes the problem of data sparseness. In this particular ill-posed problem, a reliable classification approach requires appropriate modeling of the classification process. The proposed approach is based on a hierarchical clustering algorithm in order to deal with remote sensing data in high-dimensional space. Indeed, one obvious method to perform dimensionality reduction is to use the independent component analysis process as a preprocessing step. The first particularity of our method is the special structure of its cluster tree. Most of the hierarchical algorithms associate leaves to individual clusters, and start from a large number of individual classes equal to the number of pixels; however, in our approach, leaves are associated with the most relevant sources which are represented according to mutually independent axes to specifically represent some land covers associated with a limited number of clusters. These sources contribute to the refinement of the clustering by providing complementary rather than redundant information. The second particularity of our approach is that at each level of the cluster tree, we combine both a high-level divisive clustering and a low-level agglomerative clustering. This approach reduces the computational cost since the high-level divisive clustering is controlled by a simple Boolean operator, and optimizes the clustering results since the low-level agglomerative clustering is guided by the most relevant independent sources. Then at each new step we obtain a new finer partition that will participate in the clustering process to enhance semantic capabilities and give good identification rates.

Application of independent component analysis to remove linear dependencies in EEG recorded in hippocampus.

Linear cross-spectral and nonlinear cross-bispectral analysis techniques were applied to EEG data recorded simultaneously in two subfields (CA1 and dentate gyrus) of the hippocampus in freely behaving neonatal rats during REM sleep. Linear dependencies between the two sites were specifically removed using independent component analysis (ICA). The cross-spectrum and the cross-bispectrum computed prior to ICA processing were compared with those after ICA processing to determine its effects. Our results indicate that ICA almost completely extracts the linear relationship in the theta frequency band between CA1 and the dentate gyrus. It is noted also that ICA processing significantly decreases the quadratic phase coupling (QPC) between the two hippocampal subfields.

Independent Component Ensemble of EEG for Brain–Computer Interface

Recently, successful applications of independent component analysis (ICA) to electroencephalographic (EEG) signals have yielded tremendous insights into brain processes that underlie human cognition. Many studies have further established the feasibility of using independent processes to elucidate human cognitive states. However, various technical problems arise in the building of an online brain-computer interface (BCI). These include the lack of an automatic procedure for selecting independent components of interest (ICi) and the potential risk of not obtaining a desired ICi. Therefore, this study proposes an ICi-ensemble method that uses multiple classifiers with ICA processing to improve upon existing algorithms. The mechanisms that are used in this ensemble system include: 1) automatic ICi selection; 2) extraction of features of the resultant ICi; 3) the construction of parallel pipelines for effectively training multiple classifiers; and a 4) simple process that combines the multiple decisions. The proposed ICi-ensemble is demonstrated in a typical BCI application, which is the monitoring of participants' cognitive states in a realistic sustained-attention driving task. The results reveal that the proposed ICi-ensemble outperformed the previous method using a single ICi with ∼7% (91.6% versus 84.3%) in the cognitive state classification. Additionally, the proposed ICi-ensemble method that characterizes the EEG dynamics of multiple brain areas favors the application of BCI in natural environments.

Adaptive Step-Size Natural Gradient ICA Algorithm with Weighted Orthogonalization

To improve the stability of the traditional natural gradient independent component analysis (ICA) algorithm and the accuracy of its separated results, a adaptive step-size natural gradient ICA algorithm with weighted orthogonalization is proposed. First, to take advantage of the pre-whitening pre-processing and keep the equivariance property of the ICA algorithm, based on the weighted orthogonal constraint on the separating matrix without pre-whitening of observed signals, weighted orthogonalization is introduced after the traditional gradient update. Then, according to the error estimation from the smoothed distance between separated outputs and optimal outputs, we obtain two adaptive step sizes based, respectively, on an unconstrained natural gradient ICA process and a weighted orthogonalization process. Simulation experiment results show that the speed of convergence of the adaptive step-size natural gradient ICA algorithms with weighted orthogonalization are faster than the traditional one; also, the stability of the algorithms and the accuracy of the separated results are improved observably.

ICA of fMRI Studies: New Approaches and Cutting Edge Applications

Independent component analysis (ICA) is the most commonly used and most diversely applicable exploratory method for the analysis of functional magnetic resonance imaging (fMRI) data. Over the last 10 years it has offered a wealth of insights into brain function during task execution and in the resting state. Independent component analysis is a blind source separation method that was originally applied to identify technical and physiological artifacts in fMRI, and to allow their removal prior to analysis with model-based approaches. It has matured into a method capable of offering a stand-alone assessment of activation on a sound statistical footing. Recent innovations have taken on the challenges of how components should be combined over subjects to allow group inferences, and how activation identified with ICA might be compared between groups – of patients and controls – for instance. Its reputation having been bolstered by multiple successes in the investigation of resting-state networks, ICA is being applied in other cutting edge uses of fMRI; in multivariate pattern analysis, real-time fMRI, in utero studies, with a wide variety of paradigms and stimulus types and with challenging tasks with patients at ultra-high field. These are testament both to ICA's flexibility and its evolving role both in basic neuroscience and clinical applications of fMRI. This Research Topic has attracted 19 contributions from the most renowned researchers in the field, including the inventor of Fast ICA, Aapo Hyvarinen (Hyvarinen and Ramkumar, 2013), and the authors of the most widely used ICA software for fMRI – Christian Beckmann (FSL's MELODIC) and Vince Calhoun (GIFT). The capacity of ICA to find common patterns of activation in huge cohorts of subjects is demonstrated by the parallel computing approach described by Kalcher et al. (2012) and the use of ICA with cutting edge MR methods are presented by the groups of Stefan Posse [Echo Volume Imaging (Posse et al., 2013)], Markus Barth [EEG-fMRI (Meyer et al., 2013) and Ultra-Fast Generalized Inverse Imaging (Boyacioglu et al., 2013)], and Jorge Jovicich [real-time fMRI (Soldati et al., 2013a,b)]. Two articles in this research topic reflect the continued use of ICA to identify artifacts, using the temporal characteristics of components (Rummel et al., 2013) or both temporal and spatial features (Bhaganagarapu et al., 2013). In addition to using frequency signatures to identify noise, the frequencies of signal fluctuations during rest have been studied using temporal ICA (Boubela et al., 2013) and in ultra-fast generalized imaging (Boyacioglu et al., 2013), while Di et al. (2013) examine the influence of amplitude on resting-state connectivity and Balsters et al. (2013) assess the correlation between BOLD spectral power and working memory performance. The ICA applications featured in this Research Topic range from clinical resting-state studies with patients suffering from schizophrenia (Manoliu et al., 2013; Sui et al., 2013) and neurological patients performing chin and hand motor tasks (Robinson et al., 2013) to the investigation of processing streams using chemosensory stimuli (Frasnelli et al., 2012). Combined methodological approaches are used to study belief decision making with fMRI and EEG (Douglas et al., 2013), to discriminate schizophrenia using data from fMRI, DTI, and sMRI (Sui et al., 2013), to identify amyotrophic lateral sclerosis diseased brains (Welsh et al., 2013) and to examine the microvascular specificity of the BOLD effect at 3 and 7 T using SWI (Geissler et al., 2013). We hope this collection of original research articles illustrates the extent to which ICA is becoming an increasingly flexible and potent analysis method – particularly through innovations such as real-time ICA, temporal ICA, and parallel processing implementations – and that the capacity of ICA to isolate the underlying signal sources in fMRI data is being enhanced by multimodal and ultra-fast imaging. These innovations are leading to an increase in the utility of ICA and the richness of information it can provide in both basic research work and clinical applications.

Semi-Blind Independent Component Analysis of fMRI Based on Real-Time fMRI System

Real-time functional magnetic resonance imaging (fMRI) is a type of neurofeedback tool that enables researchers to train individuals to actively gain control over their brain activation. Independent component analysis (ICA) based on data-driven model is seldom used in real-time fMRI studies due to large time cost, though it has been very popular to offline analysis of fMRI data. The feasibility of performing real-time ICA (rtICA) processing has been demonstrated by previous study. However, rtICA was only applied to analyze single-slice data rather than full-brain data. In order to improve the performance of rtICA, we proposed semi-blind real-time ICA (sb-rtICA) for our real-time fMRI system by adding regularization of certain estimated time courses using the experiment paradigm information to rtICA. Both simulated and real-time fMRI experiment were conducted to compare the two approaches. Results from simulated and real full-brain fMRI data demonstrate that sb-rtICA outperforms rtICA in robustness, computational time and spatial detection power. Moreover, in contrast to rtICA, the first component estimated by sb-rtICA tends to be the target component in more sliding windows.

An Improved Independent Component Analysis Method for Ground Targets Information Identification

The remote sensing technology can extract change information of land cover quickly, and is widely used in real-time monitoring land use changing and urban expansion. According to the characteristics of land cover information identification present in high resolution remote sensing images, a new identification algorithm of land cover change information based on independent component analysis (ICA) is proposed, which includes ICA processing, feature space optimization, and probability learning technologies. In the experiments, the land cover change information was extracted quickly using the two Google Earth images of different periods, and the extraction accuracy were evaluated. The results show that: firstly, this propose algorithm is proved to be precise, fast and easy to operate; secondly, the rate of changing to the origin areas in buildings, green fields, and roads are 98.38%, 96.57%, and 74.55%, respectively; thirdly, the urban land cover change information identification has a good image quality and it is suitable for visual and rapid mapping.

Segregation of frontoparietal and cerebellar components within saccade and vergence networks using hierarchical independent component analysis of fMRI

Cortical and subcortical functional activity stimulated via saccade and vergence eye movements were investigated to examine the similarities and differences between networks and regions of interest (ROIs). Blood oxygenation level-dependent (BOLD) signals from stimulus-induced functional Magnetic Resonance Imaging (MRI) experiments were analyzed studying 16 healthy subjects. Six types of oculomotor experiments were conducted using a block design to study both saccade and vergence circuits. The experiments included a simple eye movement task and a more cognitively demanding prediction task. A hierarchical independent component analysis (ICA) process began by analyzing individual subject data sets with spatial ICA to extract spatial independent components (sIC), which resulted in three ROIs. Using the time series from each of the three ROIs per subject, per oculomotor experiment, a temporal ICA was used to compute individual temporal independent components (tICs). For each of the three ROIs, the individual tICs from multiple subjects were entered into a second temporal ICA to compute group-level tICs for comparison. Two independent spatial maps were observed for each subject (one sIC showing activity in the frontoparietal regions and another sIC in the cerebellum) during the six oculomotor tasks. Analysis of group-level tICs revealed an increased latency in the cerebellar region when compared to the frontoparietal region. Shared neuronal behavior has been reported in the frontal and parietal lobes, which may in part explain the segregation of frontoparietal functional activity into one sIC. The cerebellum uses multiple time scales for motor learning. This may result in an increased latency observed in the BOLD signal of the cerebellar group-level tIC when compared to the frontal and parietal group-level tICs. The increased latency offers a possible explanation to why ICA dissects the cerebellar activity into an sIC. The hierarchical ICA process used to calculate group-level tICs can yield insight into functional connectivity within complex neural networks.

Independent Component Analysis for Auditory Evoked Potentials and Cochlear Implant Artifact Estimation

Auditory evoked potential (AEP) recordings have been analyzed through independent component analysis (ICA) in the literature; however, the performance varies depending on the ICA algorithms used. There are very few studies that concentrate on the optimum parameter selection for estimating the AEP components reliably, while also recovering the specific artifact generated with the normal functioning of a cochlear implant (CI). The objective of this research is to determine which ICA algorithm, high-order statistics (HOS)-based or second-order statistic (SOS)-based, is more plausible to remove this artifact and estimate the AEP. The optimal parameters of three such ICA algorithms for estimating the components from a database of recordings were determined, and then the estimates for the AEP and CI artifact were compared using each method. All the algorithms estimate the CI artifact reasonably well, although only one SOS algorithm is better positioned to estimate the AEP; this is primarily because it uses the temporal structure of this signal as part of the ICA process.

PCA and ICA processing methods for removal of artifacts and noise in electrocardiograms: A survey and comparison

Electrocardiogram (ECG) signals are affected by various kinds of noise and artifacts that may hide important information of interest. Wavelet transform (WT) technique is used to identify the characteristic points of the electrocardiogram (ECG) signal with fairly good accuracy, even in the presence of severe high frequency and low frequency noise. Independent component analysis (ICA) is a new technique suitable for separating independent components from ECG complex signals, whereas principal component analysis (PCA) is used to reduce dimensionality and for feature extraction of the ECG data prior to or at times after performing ICA in special circumstances. In this analysis, PCA is analyzed from three points of view, variance maximization, the singular value decomposition and ECG data compression. The sensitivity of the different ECG components with respect to the ECG data dimensions has been studied using PCA screen plots. The validity and performance of the approaches used are confirmed through computer simulations on common standards for electrocardiography (CSE) base ECG data. Standard or instantaneous ICA, which is the most commonly, accepted ICA technique is first compared with PCA technique and then with constrained ICA, which enables the estimation of only one component close to a particular reference ECG signal. The ICA method can also be extended for QRS detection and reference signal generation, using constrained ICA and as well for multichannel ECG separation after removing noise and artifacts, further favoring segment classification. The results were obtained using Matlab environment. Using composite WT based PCA–ICA methods helps for redundant data reduction as well for better feature extraction. The efficacy of the combined PCA–ICA algorithm lies on the fact that the location of the R-peaks is accurately determined, and none of the peaks are ignored or missed, as Quadratic Spline wavelet is also used.

Gear Multi-Faults Diagnosis of a Rotating Machinery Based on Independent Component Analysis and Fuzzy <i>K</i>-Nearest Neighbor

Gearboxes are extensively used in various areas including aircraft, mining, manufacturing, and agriculture, etc. The breakdowns of the gearbox are mostly caused by the gear failures. It is therefore crucial for engineers and researchers to monitor the gear conditions in time in order to prevent the malfunctions of the plants. In this paper, a condition monitoring and faults identification technique for rotating machineries based on independent component analysis (ICA) and fuzzy k-nearest neighbor (FKNN) is described. In the diagnosis process, the ICA was initially employed to separate characteristic vibration signal and interference vibration signal from the parallel time series obtained from multi-channel accelerometers mounted on different positions of the gearbox. The wavelet transform (WT) and autoregressive (AR) model method then were performed as the feature extraction technique to attain the original feature vector of the characteristic signal. Meanwhile, the ICA was used again to reduce the dimensionality of the original feature vector. Hence, the useless information in the feature vector could be removed. Finally, the FKNN algorithm was implemented in the pattern recognition process to identify the conditions of the gears of interest. The experimental results suggest that the sensitive fault features can be extracted efficiently after the ICA processing, and the proposed diagnostic system is effective for the gear multi-faults diagnosis, including the gear crack failure, pitting failure, gear tooth broken, compound fault of wear and spalling, etc. In addition, the proposed method can achieve higher performance than that without ICA processing with respect to the classification rate.

APPLICATION OF ICA TECHNIQUE TO PCA BASED RADAR TARGET RECOGNITION

In this paper, the ICA (independent component analysis) technique is applied to PCA (principal component analysis) based radar target recognition. The goal is to identify the similarity between the unknown and known targets. The RCS (radar cross section) signals are collected and then processed to serve as the features for target recognition. Initially, the RCS data from targets are collected by angular-diversity technique, i.e., are observed in directions of difierent elevation and azimuth angles. These RCS data are flrst processed by the PCA technique to reduce noise, and then further processed by the ICA technique for reliable discrimination. Finally, the identiflcation of targets will be performed by comparing features in the ICA space. The noise efiects are also taken into consideration in this study. Simulation results show that the recognition scheme with ICA processing has better ability to discriminate features and to tolerate noises than those without ICA processing. The ICA technique is inherently an approach of high-order statistics and can extract much important information about radar target recognition. This property will make the proposed recognition scheme accurate and reliable. This study will be helpful to many applications of radar target recognition.

Pseudo Online Independent Component Analysis for Dynamical Mixing Using Gradient Optimization

Many of the algorithms developed for independent component analysis (ICA) operate efficiently only in batch processing because they include ensemble expectation operations. The online ICA algorithm that we propose in this paper operates recursively on sequential finite lengthened inputblocksfor learning mixing parameters and for separating sources based on maximized non-Gaussianity of outputs. To enable the algorithm to work more efficiently in separating nonstationary sources and/or in dynamical mixing environments, we proposed an online estimation for the covariance matrix of input signals and an online optimization for the step size. The covariance matrix is used to calculate the whitening matrix and must be estimated appropriately for ICA processing to be efficient. Determining the step-size parameter in an online algorithm is critical to higher convergence and stability after convergence but is notoriously tiresome, especially if sources or/and the mixing environment are nonstationary. The online optimized step size we proposed is automatically optimized for suiting the powers in each block. Numerical experiments confirm that the online algorithm converges efficiently and separates sources appropriately.

A view-based statistical system for multi-view face detection and pose estimation

This study develops a novel statistical system for automatic multi-view face detection and pose estimation. The five-module detection system is based on significant local facial features (or subregions) rather than the entire face. The low- and high-frequency feature information of each subregion of the facial image are extracted and projected onto the eigenspace and residual independent basis space in order to create the corresponding PCA (principal component analysis) projection weight vector and ICA (independent component analysis) coefficient vector, respectively. Therefore, the proposed system has an improved tolerance toward different facial expressions, wide viewing angles, partial occlusions and lighting conditions. Furthermore, either projection weight vectors or coefficient vectors in the PCA or ICA space have divergent distributions and are therefore modeled by using the weighted Gaussian mixture model (GMM) rather than a single Gaussian model. The GMM weights and parameters of the GMM are estimated iteratively using the Expectation-Maximization (EM) algorithm. Face detection is then performed by conducting a likelihood evaluation process based on the estimated joint probability of the weight and coefficient vectors and the corresponding geometric positions of the subregions. The use of subregion position information can reduce the risk of false acceptances. Moreover, simple cascaded rejecter module is employed to exclude 85% of the non-face images in order to enhance the overall system performance. The computational overhead is further reduced by eliminating the requirement for a residual image reconstruction process in the ICA process. Finally, the performance of the proposed system is evaluated using challenging databases. The results not only demonstrate the ability of the system to automatically identify facial images with a high degree of accuracy, but also verify its ability to estimate the fine pose angles with 5° precision and an over 90% accuracy rate.