Wyniki wyszukiwania - BazTech

1

Infrasound Signal Classification Based on ICA and SVM

Lu Quanbo, Wang Meng, Li Mei

Archives of Acoustics

|

2023

|

Vol. 48, No. 2

191--199

EN

A diagnostic technique based on independent component analysis (ICA), fast Fourier transform (FFT), and support vector machine (SVM) is suggested for effectively extracting signal features in infrasound signal monitoring. Firstly, ICA is proposed to separate the source signals of mixed infrasound sources. Secondly, FFT is used to obtain the feature vectors of infrasound signals. Finally, SVM is used to classify the extracted feature vectors. The approach integrates the advantages of ICA in signal separation and FFT to extract the feature vectors. An experiment is conducted to verify the benefits of the proposed approach. The experiment results demonstrate that the classification accuracy is above 98.52% and the run time is only 2.1 seconds. Therefore, the proposed strategy is beneficial in enhancing geophysical monitoring performance.

2

Artifact removal from EEG signals recorded in non-restricted environmentArtifact removal from EEG signals recorded in non-restricted environment

Jamil Zainab, Jamil Afshan, Majid Muhammad

Biocybernetics and Biomedical Engineering

|

2021

|

Vol. 41, no. 2

503--515

EN

Electroencephalography (EEG) signals are always accompanied by endogenous and exogenous artifacts. Research carried out in the past few years focused on EEG artifact removal considered EEG signals recorded in a restricted lab environment. Considering the importance of EEG in daily life activities, no definitive approach is presented in removing blink artifacts from non-restricted EEG recordings. In this paper, a new supervised artifact removal method is proposed that classifies EEG chunks having eye movements and then utilizes independent component analysis and discrete wavelet transform to eliminate the ocular artifacts. The EEG data is acquired from 29 subjects in a non-restricted environment where the subject has to watch videos while walking and giving gestures and facial expressions. Thirteen morphological features are extracted from the recorded EEG signals to classify chunks with eye movements. The EEG chunks with eye movements are further processed to remove noise without distorting the morphology of signals. The proposed method is tested for eye movements and shows an improved performance in terms of correlation, mutual information, phase difference, and computational time over unsupervised modified multi-scale sample entropy and kurtosis, and wavelet enhanced independent component analysis based approaches. Moreover, the computed values of statistical parameters including sensitivity and specificity show the robustness of the proposed scheme.

3

Cooperation of CUDA and Intel multi-core architecture in the independent component analysis algorithm for EEG data

Gajos-Balińska Anna, Wójcik Grzegorz M., Stpiczyński Przemysław

Bio-Algorithms and Med-Systems

|

2020

|

Vol. 16, no. 3

art. no. 20200044

EN

Objectives: The electroencephalographic signal is largely exposed to external disturbances. Therefore, an important element of its processing is its thorough cleaning. Methods: One of the common methods of signal improvement is the independent component analysis (ICA). However, it is a computationally expensive algorithm, hence methods are needed to decrease its execution time. One of the ICA algorithms (fastICA) and parallel computing on the CPU and GPU was used to reduce the algorithm execution time. Results: This paper presents the results of study on the implementation of fastICA, which uses some multi-core architecture and the GPU computation capabilities. Conclusions: The use of such a hybrid approach shortens the execution time of the algorithm.

4

Impulse Noise Suppression Based on Power Iterative ICA in Power Line Communication

Zhang Wei, Luo Zhongqiang, Xiong Xingzhong

International Journal of Electronics and Telecommunications

|

2019

|

Vol. 65, No. 4

651--656

EN

To overcome the detrimental influence of impulse noise in power line communication and the trap of scarce prior information in traditional noise suppression schemes , a power iteration based fast independent component analysis (PowerICA) based noise suppression scheme is designed in this paper. Firstly, the pseudo-observation signal is constructed by weighted processing so that single-channel blind separation model is transformed into the multi-channel observed model. Then the proposed blind separation algorithm is used to separate noise and source signals. Finally, the effectiveness of the proposed algorithm is verified by experiment simulation. Experiment results show that the proposed algorithm has better separation effect, more stable separation and less implementation time than that of FastICA algorithm, which also improves the real-time performance of communication signal processing.

5

Automatic identifying of maternal ECG source when applying ICA in fetal ECG extraction

Yu Q., Yan H., Song L., Guo W., Liu H., Si J., Zhao Y.

Biocybernetics and Biomedical Engineering

|

2018

|

Vol. 38, no. 3

448--455

EN

Independent component analysis (ICA) is usually used as a preliminary step for maternal electrocardiogram (ECG) QRS detection in fetal ECG extraction. When applying ICA to do this, a troublesome problem arises from how to automatically identify the separated maternal ECG component. In this paper we proposed a method called PRCH (short for Peak to peak entropy, R-R interval entropy, Correlation coefficient and Heart rate) for the automatic identifying. In the method, we defined four kinds of features, including amplitude, instantaneous heart rate, morphology and average heart rate, to characterize a signal, and determined some decision parameters through machine learning. Experiments and comparison with other three existed methods were given. Through taking metric F1 for evaluation, it showed that the proposed PRCH method has the highest identifying accuracy and generalization capability.

6

Volcanic ash cloud detection from MODIS image based on CPIWS method

Liu L., Li Ch., Lei Y., Yin J., Zhao J.

Acta Geophysica

|

2017

|

Vol. 65, no. 1

151--163

EN

Volcanic ash cloud detection has been a difficult problem in moderate-resolution imaging spectroradiometer (MODIS) multispectral remote sensing application. Principal component analysis (PCA) and independent component analysis (ICA) are effective feature extraction methods based on second-order and higher order statistical analysis, and the support vector machine (SVM) can realize the nonlinear classification in low-dimensional space. Based on the characteristics of MODIS multispectral remote sensing image, via presenting a new volcanic ash cloud detection method, named combined PCA-ICA-weighted and SVM (CPIWS), the current study tested the real volcanic ash cloud detection cases, i.e., Sangeang Api volcanic ash cloud of 30 May 2014. Our experiments suggest that the overall accuracy and Kappa coefficient of the proposed CPIWS method reach 87.20 and 0.7958%, respectively, under certain conditions with the suitable weighted values; this has certain feasibility and practical significance.

7

EEG of game players - detecting involvement with and without ICA preprocessing

Górski P., Rejer I.

Przegląd Elektrotechniczny

|

2016

|

R. 92, nr 12

237--240

EN

The aim of this paper is to analyze the differences in the classification accuracy obtained with raw EEG data and with data preprocessed with Independent Components Analysis (ICA). Our main research question is whether ICA is able to improve the classification accuracy not only in the case of a multichannel recording but also when EEG data are recorded only from a few channels. In order to answer this question we performed an experiment with 6 game players and gathered EEG data during Dota 2 game session. We analyzed the EEG data separately for 19, 7, and 3 channels with and without ICA preprocessing. With all three number of channels and for each of the six players we obtained more precise classifiers, classifying the seconds of the game as involving or boring, after applying ICA (mean accuracy averaged over subjects: 19 channels - 0.87 (raw signals), 0.91 (after ICA); 7 channels - 0.8 (raw signals), 0.85 (after ICA); 3 channels - 0.75 (raw signals), 0.8 (after ICA)).

PL

Celem artykułu jest analiza różnic w dokładności klasyfikacji otrzymanej przy wykorzystaniu surowego sygnału EEG oraz sygnału poddanego preprocessingowi z wykorzystaniem analizy składowych niezależnych (ICA). Naszym głównym pytaniem badawczym jest to, czy ICA jest w stanie zwiększyć dokładność klasyfikacji nie tylko w przypadku wielokanałowego EEG, ale również wtedy, kiedy dane EEG są nagrywane tylko z kilku kanałów. W celu udzielenia odpowiedzi na to pytanie przeprowadziliśmy eksperyment z sześcioma graczami i zgromadziliśmy dane EEG podczas gry w grę Dota 2. Przeanalizowaliśmy dane oddzielnie dla 19, 7 i 3 kanałów z oraz bez zastosowania algorytmu ICA. Dla wszystkich trzech liczb kanałów i dla każdego z sześciu graczy otrzymaliśmy bardziej dokładne klasyfikatory, dokonujące klasyfikacji poszczególnych sekund gry jako angażujących i nudnych, po przeprowadzeniu preprocessingu z wykorzystaniem ICA (średnia dokładność dla wszystkich podmiotów: 19 kanałów - 0.87 (surowe sygnały), 0.91 (po ICA); 7 kanałów - 0.8 (surowe sygnały), 0.85 (po ICA); 3 kanały - 0.75 (surowe sygnały), 0.8 (po ICA)).

8

Sequential separation of twin pregnancy electrocardiograms

Kotas M., Leski J. M., Wrobel J.

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2016

|

Vol. 64, nr 1

91--101

EN

We propose to tackle the problem of maternal abdominal electric signals decomposition with a combined application of independent component analysis and projective or adaptive filtering. The developed method is employed to process the four-channel abdominal signals recorded during twin pregnancy. These signals are complicated mixtures of the maternal ECG, the ECGs of the fetal twins and noise of various origin. Although the independent component analysis cannot separate the respective signals, the proposed combination of the methods deals with this task successfully. A simulation experiment confirms high efficiency of this approach.

9

Performance comparison of ICA algorithms for audio blind source separation

Górski P.

Przegląd Elektrotechniczny

|

2015

|

R. 91, nr 2

75-78

EN

The aim of this paper is to compare five algorithms for Independent Component Analysis. The algorithms are compared with regard to performance for separating three and seven input signals. It also examined how time and number of independent components affect on separation precision. Professional sound recordings and their mixes were used for all tests.

PL

W artykule porównano pięć popularnych algorytmów z rodziny analizy składowych niezależnych. Algorytmy porównywane były pod kątem wydajności dla trzech oraz siedmiu sygnałów wejściowych. Badano również jak czas działania algorytmu oraz zwiększenie liczby składowych wejściowych wpływa na dokładność separacji. Do testów zastosowano profesjonalnie nagrane próbki śpiewu oraz ich mieszanki.

10

Inverse method for a one-stage spur gear diagnosis

Akrout A., Tounsi D., Taktak M., Abbes M. S., Haddar M.

Journal of Theoretical and Applied Mechanics

|

2015

|

Vol. 53 nr 3

617--628

EN

In this paper, a source separation approach based on the Blind Source Separation (BSS) is presented. In fact, the Independent Component Analysis (ICA), which is the main technique of BSS, consists in extracting different source signals from several observed mixtures. This inverse method is very useful in many fields such as telecommunication, signal processing and biomedicine. It is also very attractive for diagnosis of mechanical systems such as rotating machines. Generally, dynamic responses of a given mechanical system (displacements, accelerations and speeds) measured through sensors are used as inputs for the identification of internal defaults. In this study, the ICA concept is applied to the diagnosis of a one-stage gear mechanism in which two types of defects (the eccentricity error and the localized tooth defect)are introduced. The finite element method allows determination of the signals corresponding to the acceleration in some locations of the system, and those signals may be used also in the ICA algorithm. Hence, the vibratory signatures of each defect can be identified by the ICA concept. Thus, a good agreement is obtained by comparing the expected default signatures to those achieved by the developed inverse method.

11

Feature vector or time-series – comparison of gestures representations in automatic gesture recognition systems

Barczewska K., Wójtowicz W., Moszkowski T.

Challenges of Modern Technology

|

2015

|

Vol. 6, no. 1

1--5

EN

In this paper, we performed recognition of isolated sign language gestures - obtained from Australian Sign Language Database (AUSLAN) – using statistics to reduce dimensionality and neural networks to recognize patterns. We designated a set of 70 signal features to represent each gesture as a feature vector instead of a time series, used principal component analysis (PCA) and independent component analysis (ICA) to reduce dimensionality and indicate the features most relevant for gesture detection. To classify the vectors a feedforward neural network was used. The resulting accuracy of detection ranged between 61 to 87%.

12

A texture-based method for classification of schizophrenia using fMRI data

Pouyan A. A., Shahamat H.

Biocybernetics and Biomedical Engineering

|

2015

|

Vol. 35, no. 1

45--53

EN

This paper presents a texture-based method for classification of individuals into schizophrenia patient and healthy control groups based on their resting state functional magnetic resonance imaging (R-fMRI) data. In this research a combination of three different classifiers is proposed for classification of subjects into predefined groups. For all fMRI scans, the number of time points is reduced using principal component analysis (PCA) method, which projects data onto a new space. Then, independent component analysis (ICA) algorithm is used for estimation of the independent components (ICs). ICs are sorted based on their variance. For feature extraction a texture based operator called volume local binary patterns (VLBP) is applied on the estimated ICs. In order to obtain a set of features with large discrimination power, a two-sample t-test method is used. Finally, a test subject is classified into patient or control group using a combination of three different classifiers based on a majority vote method. The performance of the proposed method is evaluated using a leave-one-out cross validation method. Experimental results reveal that the proposed method has a very high accuracy.

13

Wykorzystanie analizy składowych niezależnych do automatycznej eliminacji artefaktów mrugania oczami

Król D., Kołodziej M., Majkowski A., Rak R.J.

Przegląd Elektrotechniczny

|

2014

|

R. 90, nr 8

192-194

PL

W artykule przedstawiono metodę identyfikacji i eliminacji artefaktów mrugania oczami z sygnału EEG z wykorzystaniem technik analizy składowych niezależnych i statystyk wyższych rzędów. Najistotniejszą cechą proponowanej metody jest fakt, że może ona być stosowana w sposób automatyczny, bez nadzoru użytkownika.

EN

This paper presents a method to identify and eliminate artifacts from EEG signal using independent component analysis and higher-order statistics. The key feature of the proposed method is that it can be applied in automatic manner, without user supervision.

14

Pomiary fotopulsoksymetryczne przy występowaniu zakłóceń wywołanych czynnościami ruchowymi : rozwiązania układowe

Torbicz W., Dawgul M., Dudziński K., Wawro B., Remiszewska E., Pijanowska D. G.

Elektronika : konstrukcje, technologie, zastosowania

|

2014

|

Vol. 55, nr 11

76-84

PL

Obecnie istnieje coraz szersze zapotrzebowanie na urządzenia służące do oceny stanu psychofizycznego osób, w tym żołnierzy czy sportowców, w czasie ich aktywności fizycznej, a więc w ruchu. Muszą to być przyrządy noszone. Ruch ciała jest źródłem dużych zakłóceń, które utrudniają, a nawet uniemożliwiają wykonanie pomiarów za pomocą pulsoksymetrów stosowanych w diagnostyce klinicznej. W artykule w sposób skrótowy przedstawiono zasadę pomiarów utlenowania krwi tętniczej za pomocą fotopulsoksymetrów oraz dokonano przeglądu metod przetwarzania rejestrowanych sygnałów pulsoksymetrycznych mających na celu wyeliminowanie wpływu zakłóceń ruchowych na wyniki pomiarów utlenowania krwi tętniczej.

EN

Now-a-day, there is an increasing demand for devices needed for assessing a psychophysical state of people, including soldiers and sportsmen, during their physical activity, and so in moving. Such devices must be worn. Body movement is a source of high disturbances, which impede or even make impossible realization of measurements by pulse oximeters applied in clinical diagnosis. The paper briefly presents basic information on arterial blood oxygen saturation measurements using pulse oximeters, and gives a review of methods used for processing the monitored pulse oximeter signals in order to eliminate an influence of the movement disturbances on the results of oxygen arterial blood saturation measurements.

15

Shear wave velocity prediction using seismic attributes and well log data

Gholami R., Moradzadeh A., Rasouli V., Hanachi J.

Acta Geophysica

|

2014

|

Vol. 62, no. 4

818--848

EN

Formation’s properties can be estimated indirectly using joint analysis of compressional and shear wave velocities. Shear wave data is not usually acquired during well logging, which is most likely for cost saving purposes. Even if shear data is available, the logging programs provide only sparsely sampled one-dimensional measurements: this information is inadequate to estimate reservoir rock properties. Thus, if the shear wave data can be obtained using seismic methods, the results can be used across the field to estimate reservoir properties. The aim of this paper is to use seismic attributes for prediction of shear wave velocity in a field located in southern part of Iran. Independent component analysis (ICA) was used to select the most relevant attributes to shear velocity data. Considering the nonlinear relationship between seismic attributes and shear wave velocity, multi-layer feed forward neural network was used for prediction of shear wave velocity and promising results were presented.

16

Wykorzystanie metody PCA i ICA do analizy sygnału EEG w kontekście usuwania zakłóceń

Paszkiel S.

Pomiary Automatyka Kontrola

|

2013

|

R. 59, nr 3

204--207

PL

W artykule przedstawiono metodę PCA (ang. Principal Component Analysis) oraz ICA (ang. Independent Component Analysis), jako narzędzia pomocne w procesie eliminacji artefaktów z sygnału elektroencefalograficznego. Proces rejestracji sygnału elektroencefalograficznego można zobrazować, jako BSS (ang. Blind Signals Separation). Dzięki temu możliwe jest dokonywanie estymacji nieznanych sygnałów źródłowych oraz ekstrakcji niepożądanych sygnałów zakłócających, w zakresie ich późniejszej eliminacji. W tym celu konieczne jest doskonalenie metod weryfikacji i eliminacji artefaktów z sygnału EEG. W artykule opisano możliwość zastosowania powyższych metod w zakresie sygnału EEG oraz zrealizowane zostało porównanie skuteczności ich działania.

EN

: In the paper there are presented the Principal Component Analysis (PCA) and the Independent Component Analysis (ICA) as useful tools for elimination of artefacts in an electroencephalographic signal. The process of registration of the electroencephalographic signal can be described as BSS - Blind Signals Separation. It is possible to estimate unknown source signals and to extract intrusive disturbing signals in terms of their subsequent elimination. It is necessary to improve the methods of verification and elimination of artefacts from an EEG signal. The Brain Computer Interface (BCI) technology is presented briefly in the first part of the paper. EEG signal characteristics and its acqui-sition with the non-invasive method are described in the second part. Next, there is discussed the possibility of using the PCA and ICA methods in terms of analysis of an EEG signal. Comparison of the effectiveness of these methods is presented as well. A general profile of the EEG signal processing is shown in Fig. 1. An example of use of the infomax algorithm for a real EEG signal is depicted in Fig. 2. Fig. 3 shows an exemplary Event-Related Potential (ERP) of the EEG signal.

17

Direct assessment of the fetal heart rate from abdominal composite recordings

Moslem B., Bazzi O., Khalaf J., Diab M.O.

Journal of Medical Informatics & Technologies

|

2013

|

Vol. 22

143--149

EN

In respect to the main goal of our ongoing work for analyzing fetal electrocardiogram (FECG) signals for monitoring the health of the fetus, we investigate in this paper the possibility of extracting the fetal heart rate (FHR) directly from the abdominal composite recordings. Our proposed approach is based on a combination of Independent Component Analysis (ICA) and least mean square (LMS) adaptive filter. The FHR of the estimated FECG signal is finally compared to a reference value extracted from a FECG signal recorded by using a spiral electrode attached directly to the fetal scalp. The experimental results show that FHR can be successfully evaluated directly from the abdominal composite recordings without the need of using any external reference signal.

18

Multidimensional independent subspace analysis by natural gradient

Wang F., Zhang L., Li R., Zhou Y., Liu N.

Przegląd Elektrotechniczny

|

2012

|

R. 88, nr 9b

51-54

EN

Multidimensional Independent Subspace Analysis (MISA) as an extended Independent Component Analysis (ICA) method has been considered. The general and detailed definition, existence, uniqueness, separability of the MISA model are given and the relationships between ICA and MISA are also discussed. The natural gradient separation algorithm and corresponding simulation results for MISA are constructed based on the maximum likelihood theory and natural gradient method.

PL

W artykule zaprezentowano metodę MISA – multidimensional independent subspace analysis. Przedstawiono też metode IOCA – independent component analysis. Opracowano algorytm separacji – natural gradient separation algorithm.

19

Removing physiological artifacts from the EEG data by algorithms based on differential entropy

Górecka J.

Pomiary Automatyka Kontrola

|

2012

|

R. 58, nr 11

975-977

EN

A new form of the nonlinearity implemented in the ICA approach is presented in the paper. The proposed independent component analysis based on differential entropy can be used for elimination of physiological artifacts from electroencephalographic signals. For verification of the quality of separation of the EEG data, the PI index is proposed. The second measure of accuracy is the normalized kurtosis which can be used in analysis of the simulated EEG data. As it has been proved, the new sigmoid function used in the ICA approach can effectively separate the EEG data.

PL

W artykule przedstawiono nową propozycję nieliniowości - sigmoidalną funkcję algebraiczną, która została zaimplementowana w algorytmie stosującym metodę analizy składowych niezależnych (ang. Independent Component Analysis). Proponowana nowa postać algorytmu wykorzystująca właściwości entropii różniczkowej, może zostać użyta także do separacji a następnie eliminacji wybranych artefaktów fizjologicznych pochodzenia ocznego i mięśniowego zarejestrowanych w zapisach EEG. W celu weryfikacji dokładności separacji sygnałów EEG zaproponowano współczynnik jakości separacji PI (ang. Performance Index). Jako drugą miarę dokładności procesu separacji wybrano wartość znormalizowanej kurtozy, która może być stosowana jedynie w przypadku separacji elektroencefalogramów zarejestrowanych z symulatora EEG. W artykule udowodniono, że użycie nowej funkcji sigmoidalnej w rozszerzonej postaci algorytmu infomax prowadzi do efektywnej separacji sygnałów EEG umożliwiając eliminację wybranych składowych niepożądanych.

20

Analysis of complex-valued functional magnetic resonance imaging data: are we just going through a "phase"?

Calhoun V., Adali T.

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2012

|

Vol. 60, nr 3

371-418

EN

Functional magnetic resonance imaging (fMRI) data are acquired as a natively complex data set, however for various reasons the phase data is typically discarded. Over the past few years, interest in incorporating the phase information into the analyses has been growing and new methods for modeling and processing the data have been developed. In this paper, we provide an overview of approaches to understand the complex nature of fMRI data and to work with the utilizing the full information, both the magnitude and the phase. We discuss the challenges inherent in trying to utilize the phase data, and provide a selective review with emphasis on work in our group for developing biophysical models, preprocessing methods, and statistical analysis of the fully-complex data. Of special emphasis are the use of data-driven approaches, which are particularly useful as they enable us to identify interesting patterns in the complex-valued data without making strong assumptions about how these changes evolve over time, something which is challenging for magnitude data and even more so for the complex data. Finally, we provide our view of the current state of the art in this area and make suggestions for what is needed to make efficient use of the fully-complex fMRI data.