Independent Component Analysis Feature Extraction Research Articles

Preservation of EEG spectral power features during simultaneous EEG-fMRI.

Electroencephalographic (EEG) data quality is severely compromised when recorded inside the magnetic resonance (MR) environment. Here we characterized the impact of the ballistocardiographic (BCG) artifact on resting-state EEG spectral properties and compared the effectiveness of seven common BCG correction methods to preserve EEG spectral features. We also assessed if these methods retained posterior alpha power reactivity to an eyes closure-opening (EC-EO) task and compared the results from EEG-informed fMRI analysis using different BCG correction approaches. Electroencephalographic data from 20 healthy young adults were recorded outside the MR environment and during simultaneous fMRI acquisition. The gradient artifact was effectively removed from EEG-fMRI acquisitions using Average Artifact Subtraction (AAS). The BCG artifact was corrected with seven methods: AAS, Optimal Basis Set (OBS), Independent Component Analysis (ICA), OBS followed by ICA, AAS followed by ICA, PROJIC-AAS and PROJIC-OBS. EEG signal preservation was assessed by comparing the spectral power of traditional frequency bands from the corrected rs-EEG-fMRI data with the data recorded outside the scanner. We then assessed the preservation of posterior alpha functional reactivity by computing the ratio between the EC and EO conditions during the EC-EO task. EEG-informed fMRI analysis of the EC-EO task was performed using alpha power-derived BOLD signal predictors obtained from the EEG signals corrected with different methods. The BCG artifact caused significant distortions (increased absolute power, altered relative power) across all frequency bands. Artifact residuals/signal losses were present after applying all correction methods. The EEG reactivity to the EC-EO task was better preserved with ICA-based correction approaches, particularly when using ICA feature extraction to isolate alpha power fluctuations, which allowed to accurately predict hemodynamic signal fluctuations during the EEG-informed fMRI analysis. Current software solutions for the BCG artifact problem offer limited efficiency to preserve the EEG spectral power properties using this particular EEG setup. The state-of-the-art approaches tested here can be further refined and should be combined with hardware implementations to better preserve EEG signal properties during simultaneous EEG-fMRI. Existing and novel BCG artifact correction methods should be validated by evaluating signal preservation of both ERPs and spontaneous EEG spectral power.

Gear fault diagnosis method based on wavelet-packet independent component analysis and support vector machine with kernel function fusion

Aiming at the problem of gear fault diagnosis, in order to effectively extract the features and improve the accuracy of gear fault diagnosis, the method based on wavelet-packet independent component analysis and support vector machine with kernel function fusion is proposed in this research. The proposed wavelet-packet independent component analysis feature extraction method can effectively combine the advantages of wavelet packet and independent component analysis methods and acquire more comprehensive feature information. Besides, the proposed kernel-function-fusion support vector machine can well integrate the advantage characteristics of each kernel function. The energy features of wavelet packet coefficients are acquired with four-layer wavelet packet decomposition and then the extracted energy features are further optimized by the independent component analysis method. The kernel-function-fusion support vector machine method is adopted to realize the gear fault diagnosis. Two kernel function models with the best self-classification accuracy are employed to serve the gear fault diagnosis corporately. The test samples are primarily classified by the main kernel function model, and then some samples are selected to be reclassified with the other kernel function model. Finally, the two kernel function models cooperate to determine the type of test samples. The comparison investigations demonstrate that the proposed method based on wavelet-packet independent component analysis and support vector machine with kernel function fusion achieves very high diagnosis accuracy.

Modeling Method Based on Smooth Support Vector Regression with Independent Component Analysis Feature Extraction

Smooth Support Vector Regression (SSVR) is new modified edition of traditional support vector regression for better performance. To further improve the modeling capability of SSVR, it is necessary to take into account the feature extraction based on Independent Component Analysis (ICA) before SSVR. Simulation on the example of function approximation shows that the result of SSVR based on ICA feature extraction is better than that of SSVR without ICA preprocess.

Research on RBFNN Modeling Based on ICA Feature Extraction

In order to accurately depict the complicated characteristics of nonlinear system by modeling, a Radial Basis Function Network (RBFNN) modeling method based on Independent Component Analysis (ICA) is proposed. First ICA is perform for extracting basic features of the training samples, and then the extracted basic features is used to establish to RBFNN model. The simulation indicates that, the hybrid modeling method proposed is better than that of another 2 methods with simple model structure, and is effective and feasible to establish for the nonlinear modeling system.

Defect Detection in Solar Modules Using ICA Basis Images

Solar power has become an attractive alternative of electricity energy. Solar cells that form the basis of a solar power system are mainly based on multicrystalline silicon. A set of solar cells are assembled and interconnected into a large solar module to offer a large amount of electricity power for commercial applications. Many defects in a solar module cannot be visually observed with the conventional CCD imaging system. This paper aims at defect inspection of solar modules in electroluminescence (EL) images. The solar module charged with electrical current will emit infrared light whose intensity will be darker for intrinsic crystal grain boundaries and extrinsic defects including micro-cracks, breaks and finger interruptions. The EL image can distinctly highlight the invisible defects but also create a random inhomogeneous background, which makes the inspection task extremely difficult. The proposed method is based on independent component analysis (ICA), and involves a learning and a detection stage. The large solar module image is first divided into small solar cell subimages. In the training stage, a set of defect-free solar cell subimages are used to find a set of independent basis images using ICA. In the inspection stage, each solar cell subimage under inspection is reconstructed as a linear combination of the learned basis images. The coefficients of the linear combination are used as the feature vector for classification. Also, the reconstruction error between the test image and its reconstructed image from the ICA basis images is also evaluated for detecting the presence of defects. Experimental results have shown that the image reconstruction with basis images distinctly outperforms the ICA feature extraction approach. It can achieve a mean recognition rate of 93.4% for a set of 80 test samples.

Detection of signal transients using independent component analysis and its application in gearbox condition monitoring

This paper addresses feature extraction of the higher-order statistics, which can effectively characterize the transients, using independent component analysis (ICA) for the one-dimensional measured vibration signal, and then proposes a novel automatic technique for detecting the transients in vibration signals with the low signal-to-noise ratio by ICA feature extraction. The basic principle of the ICA-based transient detection method is that the independent components (ICs) coefficients of the transients and the noise can be effectively distinguished by their different sparseness properties. Specifically, the proposed method mainly includes three steps: training the ICA basis features from the signal segments, denoising the sparse ICs coefficients using the shrinkage function deduced by the maximum a posteriori (MAP) estimation, and reconstructing the transient segments by the shrunken coefficients through the ICA basis functions. Experimental results through the simulated signal analysis and the vibration signal analysis show that the ICA-based method is very effective for transient detection outperforming the traditional methods and is valuable for gearbox condition monitoring and fault diagnosis.

A comparison of PCA, KPCA and ICA for dimensionality reduction in support vector machine

Recently, support vector machine (SVM) has become a popular tool in time series forecasting. In developing a successful SVM forecastor, the first step is feature extraction. This paper proposes the applications of principal component analysis (PCA), kernel principal component analysis (KPCA) and independent component analysis (ICA) to SVM for feature extraction. PCA linearly transforms the original inputs into new uncorrelated features. KPCA is a nonlinear PCA developed by using the kernel method. In ICA, the original inputs are linearly transformed into features which are mutually statistically independent. By examining the sunspot data, Santa Fe data set A and five real futures contracts, the experiment shows that SVM by feature extraction using PCA, KPCA or ICA can perform better than that without feature extraction. Furthermore, among the three methods, there is the best performance in KPCA feature extraction, followed by ICA feature extraction.