Classification on the Visual Evoked Potential Based on Wavelet and BP Network

段广强,邹俊忠,张见

DOI: https://doi.org/10.3969/j.issn.1674-1242.2009.01.002

2009-01-01

Abstract:In this paper,Wavelet decomposition and BP networks are used to identify Visual Evoked Potential. First,Wavelet decomposition is used to extract feature and reduce dimension of data.Second,BP network is used to classification.

What problem does this paper attempt to address?

EEG Recognition Based on Multiple Types of Information by Using Wavelet Packet Transform and Neural Networks

Baojun Wang,Liu Jun,Jing Bai,Le Peng,Yan Li,Guang Li

DOI: https://doi.org/10.1109/iembs.2005.1615697

2005-01-01

Abstract:In this work, we proposed a method for a binary classification in an EEG-based brain computer interface (BCI) with wavelet packet transform and neural networks. For feature extraction, we introduced a new method which combined the slow cortical potentials (SCPs) and the specific energy from the time-frequency domain in beta-band via the wavelet packet transform. A 3-layer perceptron established by back-propagation and the support vector machines (SVMs) were utilized for classification, respectively. We compared the performance in terms of changing the architecture of the net. The accuracy of BP network was found to be best with 4-5-1 architecture reaching an accuracy of 91.47% on test set. Meanwhile, a SVM with Gaussian kernel revealed an accuracy of 91.13% on the test set, showing that multiple types of information have great advantage over other features
Extraction And Classification Of Visual Evoked Potentials Based On A Two-Stage Source Extraction Algorithm

Xiuling Wu,Liqing Zhang,Zhilin Zhang,Wenjun Zhu

DOI: https://doi.org/10.1109/ICCIAS.2006.295333

2006-01-01

Abstract:In order to verify whether or not the EEG patterns can be classified when the subjects perceive different types of geometric figures, we perform some EEG experiments. In this paper the evoked potentials by three types of geometric figures are extracted and classified using a series of approaches. First, a two-stage source extraction algorithm is proposed to extract the evoked potentials from the recorded EEG signals, and then a mutual information based feature selection method is presented to find effective features for classification. Finally, a multi-category support vector machine classifier is employed, which achieves the average classification performance of 93.2%.
Pattern Classification of Visual Evoked Potentials Based on Parallel Factor Analysis

Jie Li,Liqing Zhang,Qibin Zhao

DOI: https://doi.org/10.1007/978-1-4020-8387-7_99

2008-01-01

Abstract:Visual Evoked Potentials (VEPs) reflect the brain’s mental process to specific stimuli, including perception and recognition. Feature analysis of VEPs evoked by geometric figures is of significance in understanding visual neural mechanism and has potential applications in the field of brain computer interface and biomedical engineering. We use Parallel Factor (PARAFAC) model to extract features of the VEPs triggered by three classes of geometric figures, and construct the computational model for class discrimination. PARAFAC is used to decompose the wavelet transformed VEPs. Then by the proposed computational model, we can project single trial data into the subspace spanned by channel× frequency× time of the factors to obtain the feature vectors. We further use SVM to classify the feature vectors of the selected two classes, achieving the highest classification accuracy 80%.
A Method of the Identification of Ventricular Late Potentials

娄智,杨基海,陈香,钟展辉,梁政,钱晓进,郑凡

DOI: https://doi.org/10.3969/j.issn.0253-2778.2002.02.015

2002-01-01

JUSTC

Abstract:The identification of Ventricular Late Potentials (VLP) is completed by means of a method combining wavelet packet analysis with artificial neural network in this paper. At first, the eigenvector of ECG signals, which comprises energy of sub-signals in different frequency bands, is extracted by using the technique of wavelet packet decomposition. Then, the classification of ECG signals are realized by adopting Radial Basis Function Network (RBFN). The experiment data obtained from 28 examples of signal average ECG with three leads are processed. The result showe that the method proposed in this paper is quite satisfying for the identification of VLPs.
An Attention-Based Wavelet Convolution Neural Network for Epilepsy EEG Classification

Qi Xin,Shaohai Hu,Shuaiqi Liu,Ling Zhao,Yu-Dong Zhang

DOI: https://doi.org/10.1109/tnsre.2022.3166181

IF: 4.9

2022-01-01

IEEE Transactions on Neural Systems and Rehabilitation Engineering

Abstract:As a kind of non-invasive, low-cost, and readily available brain examination, EEG has attached significance to the means of clinical diagnosis of epilepsy. However, the reading of long-term EEG records has brought a heavy burden to neurologists and experts. Therefore, automatic EEG classification for epileptic patients plays an essential role in epilepsy diagnosis and treatment. This paper proposes an Attention Mechanism-based Wavelet Convolution Neural Network for epilepsy EEG classification. Attention Mechanism-based Wavelet Convolution Neural Network firstly uses multi-scale wavelet analysis to decompose the input EEGs to obtain their components in different frequency bands. Then, these decomposed multi-scale EEGs are input into the Convolution Neural Network with an attention mechanism for further feature extraction and classification. The proposed algorithm achieves 98.89% triple classification accuracy on the Bonn EEG database and 99.70% binary classification accuracy on the Bern-Barcelona EEG database. Our experiments prove that the proposed algorithm achieves a state-of-the-art classification effect on epilepsy EEG.

engineering, biomedical,rehabilitation
[Feature Extraction and Classification of EEG for Mental Tasks Based on Wavelet Packet Analysis].

Jianzhong Xue,Weixing He,Xiangguo Yan,Chongxun Zheng

DOI: https://doi.org/10.3321/j.issn:1001-5515.2004.03.014

2004-01-01

Journal of Biomedical Engineering

Abstract:This paper explores the use of wavelet packet analysis to extract features from spontaneous electroencephalogram (EEG) during three different mental tasks. Artifact-free EEG segments are transformed to multi-scale representations by dyadic wavelet packet decomposition channel by channel. Their feature vectors formed by energy values of different sub-spaces EEG components are used as inputs of a radial basis function network to test the classification accuracies of three task pairs. The results indicate that the classification accuracies of the wavelet packet analysis method are significantly better than those of autoregressive model method. Wavelet packet analysis would be a promising method to extract features from EEG signals.
P300 feature extraction algorithm based on wavelet transform and temporal energy entropy

Wang Pan,Shen Jizhong,Shi Jinhe

DOI: https://doi.org/10.19650/j.cnki.cjsi.2011.06.013

2011-01-01

Abstract:Aiming at the drawback of slow training and testing speed in P300 feature extraction and classification, a new P300 feature extraction method based on P300 in band and out of band EEG signal characteristics was proposed, which combines wavelet transform with temporal energy entropy. The proposed method overcomes the harsh requirement for electrode number and number of averaging. In this paper, support vector machine is used as classifier and the P300 data sets of BCI Competition 2003 and BCI Competition 2005 are used to verify the method. Verification results show that the proposed method uses only one electrode data and 2 times signal averaging, and still can obtain very good classification result and consume short computation time.
Wavelet Packet Transformation Feature Extraction and Surface EMG Signal Classification

谢洪波,王志中,黄海

DOI: https://doi.org/10.3969/j.issn.1003-8868.2003.09.004

2003-01-01

Abstract:A surface electromyography (SEMG) signal classification method based on wavelet packet transformation (WPT) is presented in this paper. The feature extraction method is analyzed. The energies in different frequency bands selected as robust feature vectors, four types of forearm movement are identified through learning vector quantization neural network. Compared with other time-frequency analysis method, this method has a higher identification rate and great potential in analyzing other non-stationary physiological signals.
BP neural network based on principle component analysis in multi-spectral remote sensing images classification

HU Jian-ce,WU Guo-ping

2009-01-01

Abstract:In order to improve the classification accuracy of multi-spectral remote sensing image,this paper puts forward a new classification method based on principal component analysis.The method is consisted of two steps: reducing the dimensions of multi-spectral remote sensing image with principle component analysis and generating a new image by the three main components of the remote sensing image;performing supervised classification on the new image with BP neural network.The result indicates that this method is superior to traditional algorithms,and its overall accuracy and Kappa coefficient reach 88.5% and 0.862.
Weighted Hermite Variable Projection Networks for Classifying Visually Evoked Potentials

Tamas Dozsa,Carl Bock,Jens Meier,Peter Kovacs

DOI: https://doi.org/10.1109/TNNLS.2024.3475271

2024-10-28

Abstract:The occipital cortex responds to visual stimuli regardless of a patient's level of consciousness or attention, offering a noninvasive diagnostic tool for both ophthalmologists and neurologists. This response signal manifests as a unique waveform referred to as the visually evoked potential (VEP), which can be extracted from the electroencephalogram (EEG) activity of a human being. We propose a trainable VEP representation to disentangle the underlying explanatory factors of the data. To enhance the learning process with domain knowledge, we present an innovative parameterization of classical Hermite functions that effectively captures VEP pattern variations arising from patient-specific factors, disorders, and measurement setup influences. Then, we introduce a differentiable variable projection (VP) layer to fuse Hermite basis function expansions (BFEs) of VEP signals with machine learning (ML) approaches. We prove the existence of an optimal set of parameters in the least-squares sense, assess the representation power of such layers, and calculate their analytical derivatives, which allows us to utilize backpropagation for training. Finally, we evaluate the effectiveness of the proposed learning framework in VEP-based color classification. To achieve this, we have designed a novel measurement system dedicated to intraoperative clinical use cases, which presents new ways for patient monitoring during neurosurgical procedures.
EEG Feature Extraction Based on Wavelet Packet Decomposition for Brain Computer Interface

Yang Bang-hua,Yan Guo-zheng,Yan Rong-guo,Wu Ting

DOI: https://doi.org/10.1016/j.measurement.2007.07.007

2007-01-01

Abstract:A method based on wavelet packet best basis decomposition (WPBBD) is investigated for the purpose of extracting features of electroencephalogram signals produced during motor imagery tasks in brain-computer interfaces. The method includes the following three steps. (1) Original signals are decomposed by wavelet packet transform (WPT) and a wavelet packet library can be formed. (2) The best basis for classification is selected from the library. (3) Subband energies included in the best basis are used as effective features. Three different motor imagery tasks are discriminated using the features. The WPBBD produces a 70.3% classification accuracy, which is 4.2% higher than that of the existing wavelet packet method.
Steady-State Visual Evoked Potential Classification Using Complex Valued Convolutional Neural Networks

Akira Ikeda,Yoshikazu Washizawa

DOI: https://doi.org/10.3390/s21165309

IF: 3.9

2021-08-06

Sensors

Abstract:The steady-state visual evoked potential (SSVEP), which is a kind of event-related potential in electroencephalograms (EEGs), has been applied to brain–computer interfaces (BCIs). SSVEP-based BCIs currently perform the best in terms of information transfer rate (ITR) among various BCI implementation methods. Canonical component analysis (CCA) or spectrum estimation, such as the Fourier transform, and their extensions have been used to extract features of SSVEPs. However, these signal extraction methods have a limitation in the available stimulation frequency; thus, the number of commands is limited. In this paper, we propose a complex valued convolutional neural network (CVCNN) to overcome the limitation of SSVEP-based BCIs. The experimental results demonstrate that the proposed method overcomes the limitation of the stimulation frequency, and it outperforms conventional SSVEP feature extraction methods.

engineering, electrical & electronic,chemistry, analytical,instruments & instrumentation
A CWT-based SSVEP classification method for brain-computer interface system

Zimu Zhang,Xiuquan Li,Zhidong Deng

DOI: https://doi.org/10.1109/ICICIP.2010.5564336

2010-01-01

Abstract:Continuous wavelet transform (CWT) is used in this paper for steady-state visual evoked potential (SSVEP) detection in a brain-computer interface (BCI) system. The developed BCI system is designed for the remote control of humanoid robot through wireless sensor networks (WSN). A new CWT-based feature extraction method is presented and the whole framework of the BCI system is described. We investigated the feature extraction perfomance for different kinds of mother wavelets. Performance comparison was also conducted between CWT and fast Fourier transform (FFT). The experimental results show that the CWT-based method outperforms the FFT-based one in the SSVEP feature extraction scheme, specifically for short EEG segments. Moreover, the complex Morlet wavelet has significant superiority over several other mother wavelets in the CWT-based SSVEP feature extraction.
Face recognition fellow-up classification algorithm based on BP neural network

Zhou WEN,Xiao-wei SHAO,De-ren GONG

2011-01-01

Journal of Computer Applications

Abstract:Back-Propagation Neural Network(BPNN) was used in the face recognition in the paper.How to accurately classify and recognize these outputs which are obtained under the BPNN model,three different fellow-up classification approaches were used to classify obtained outputs.The first method,which is widely used,is maximum-component value recognition based on output vector;the next is threshold classification based on all components of output vector;the third approach,based on N-dimensional vector space,is central area classification.Experimental results through the face data of Auto Targets Recognition(ATR) show that the last two algorithms are feasible and effective for face recognition,and the error recognizing-accuracy is down to 2.2%,rejecting-accuracy is up to 93.21%.
A PVC Identification Method of ECG Signal Based on Improved BPNN

Xiong Yang,Xin Yu Jin,Jian Feng Shen

DOI: https://doi.org/10.4028/www.scientific.net/amm.738-739.578

2015-01-01

Applied Mechanics and Materials

Abstract:Computer-aided diagnosis of Premature Ventricular Contraction (PVC) plays an important role in timely detection and treatment of arrhythmias. Conventional identification methods based on back propagation neural network (BPNN) get problems of overlong training time and local optimum. This paper proposes an application of improved BPNN on PVC identification and the improvements of BPNN are based on self-adaptive learning rate and momentum in training. Denoising and feature extraction of ECG signal obtained from MIT-BIH arrhythmia database are processed first. A comparison between standard BPNN and improved BPNN shows that the latter gets less training time and better accuracy.
A comparing network for the classification of steady-state visual evoked potential responses based on convolutional neural network

Jiczhen Xing,Shuang Qiu,Chenyao Wu,Xuelin Ma,Jinpeng Li,Huiguang He

DOI: https://doi.org/10.1109/CIVEMSA45640.2019.9071633

2019-01-01

Abstract:Brain-computer interfaces (BCIs) based on Steady-State Visual Evoked Potentials (SSVEPs) has been attracting much attention because of its high information transfer rate and little user training. However, most methods applied to decode SSVEPs are limited to CCA and some extended CCA-based methods. This study proposed a comparing network based on Convolutional Neural Network (CNN), which was used to learn the relationship between EEG signals and the templates corresponding to each stimulus frequency of SSVEPs. This novel method incorporated prior knowledge and a spatial filter (task related component analysis, TRCA) to enhance detection of SSVEPs. The effectiveness of the proposed method was validated by comparing it with the standard CCA and other state-of-the art methods for decoding SSVEPs (i.e., CNN and TRCA) on the actual SSVEP datasets collected from 17 subjects. The comparison results indicated that the CNN-based comparing network significantly could significantly improve the classification accuracy compared with the standard CCA, TRCA and CNN. Furthermore, the comparing network with TRCA achieved the best performance among three methods based on comparing network with the averaged accuracy of 84.57% (data length: 2s) and 70.21% (data length: 1s). The study validated the efficiency of the proposed CNN-based comparing methods in decoding SSVEPs. It suggests that the comparing network with TRCA is a promising methodology for target identification of SSVEPs and could further improve the performance of SSVEP-based BCI system.
Automatic Classification of Epileption Activity in EEG Based on Wavelet Analysis and Teager Energy Operator

Zou Junzhong,Zhang Jian

DOI: https://doi.org/10.3969/j.issn.1674-1242.2012.03.002

2012-01-01

Progress in Biomedical Engineering

Abstract:This paper proposes an epileptic EEG classification mathod that applies the wavelet transformation the nonlinear energy operator(NEO) and the approximate entropy to distinct the differences between health EEG and epileptic EEG with SVM.Firstly,the EEG signals were decomposed up to 4 levels using wavelet analysis. Then,the NEOs of those detailed coefficients with frequencies being closed to spike were calculated,After that,the obtained NEOs were performed approximate entropy computation.At last,EEG signals were classified with SVM.Experimental results showed that the system performed well for classifying both epilepsy and normal EEG signals.
InceptionSSVEP: A Multi-Scale Convolutional Neural Network for Steady-State Visual Evoked Potential Classification

Yipeng Du,Mingxi Yin,Bingli Jiao

DOI: https://doi.org/10.1109/iccc51575.2020.9345194

2020-01-01

Abstract:This paper presents a deep learningbased classification model, referred to as InceptionSSVEP, for the steady-state visual evoked potential (SSVEP) based braincomputer interface (BCI). InceptionSSVEP adopts the main concept of Inception network, which is a deep learning model performing well in image classification tasks, to improve the performance of SSVEP classification. A multi-scale convolution structure is utilized in InceptionSSVEP to extract both long-term and short-term features of SSVEP signals, for the purpose of ensuring the comprehensiveness of high-dimensional features in extracted SSVEPs. Moreover, a data enhancement scheme is proposed to overcome the impact of SSVEP data amount limitation on classifier training. Results show that the proposed InceptionSSVEP outperforms other existing methods significantly, and validate that Inception networks have good transferability on SSVEP signals. Reasons for the good performance of InceptionSSVEP are analyzed using deep learning interpretability tools.
Circuit Breaker Fault Analysis Based on Wavelet Packet Time - Frequency Entropy and LM Algorithm to Optimize BP Neural Network

Zijian Guo,Mingliang Liu,Bing Li

DOI: https://doi.org/10.23919/chicc.2018.8483242

2018-01-01

Abstract:High-voltage circuit breakers are the most important control and protection measures in power systems, and their reliable operation is critical to the safety and stability of power systems. However, the high-voltage circuit breaker machinery often fails, the vibration signal of the high-voltage circuit breaker hides the rich fault information. The change of the vibration signal reflects the mechanical state of the circuit breaker, and the vibration vector feature vector extraction and classification is the most important problem in fault diagnosis. In this paper, a BP neural network and wavelet packet time - frequency entropy method based on LM optimization algorithm are proposed. The vibration signal of the circuit breaker is extracted and faulted. The vibration signal of the high voltage circuit breaker is decomposed by wavelet packet, and then the time-frequency entropy of the vibration signal is obtained as the eigenvector. The feature vector is input to the BP neural network optimized by LM to determine the working state and fault type of the circuit breaker. Experiments show that the BP neural network optimized by LM algorithm and wavelet packet time-frequency entropy can be used to judge the mechanical failure of high voltage side circuit breaker more efficiently.
Feature Extraction of EEG for Imagery Left-Right Hands Movement

王攀,沈继忠,施锦河

DOI: https://doi.org/10.3969/j.issn.1004-1699.2010.09.003

2010-01-01

Abstract:With the aim to solve the problems in brain-computer interfaces such as huge amounts of features, heavy computation and low classification accuracy in the traditional methods for the feature extraction of electroencephalography(EEG), a new method based on time domain, frequency domain and space domain was proposed, which could extract event related desynchronization and event related synchronization(ERD/ERS) signals during imagining left and right hands movement. In this paper, Independent Component Analysis(ICA) and wavelet transform was used to extract the temporal, spectral and spacial features from the original EEG signals, and then the extracted features were classified by the BP(Back Propagation) neural network. The classification results showed that the features of EEG, which were extracted during imagining left and right hands movement with the proposed method, could effectively overcome the drawbacks of the traditional method based solely on time-frequency domain when describing the characteristic of the brain electrical signals. And the proposed method displayed better classification performance.

Classification on the Visual Evoked Potential Based on Wavelet and BP Network

EEG Recognition Based on Multiple Types of Information by Using Wavelet Packet Transform and Neural Networks

Extraction And Classification Of Visual Evoked Potentials Based On A Two-Stage Source Extraction Algorithm

Pattern Classification of Visual Evoked Potentials Based on Parallel Factor Analysis

A Method of the Identification of Ventricular Late Potentials

An Attention-Based Wavelet Convolution Neural Network for Epilepsy EEG Classification

[Feature Extraction and Classification of EEG for Mental Tasks Based on Wavelet Packet Analysis].

P300 feature extraction algorithm based on wavelet transform and temporal energy entropy

Wavelet Packet Transformation Feature Extraction and Surface EMG Signal Classification

BP neural network based on principle component analysis in multi-spectral remote sensing images classification

Weighted Hermite Variable Projection Networks for Classifying Visually Evoked Potentials

EEG Feature Extraction Based on Wavelet Packet Decomposition for Brain Computer Interface

Steady-State Visual Evoked Potential Classification Using Complex Valued Convolutional Neural Networks

A CWT-based SSVEP classification method for brain-computer interface system

Face recognition fellow-up classification algorithm based on BP neural network

A PVC Identification Method of ECG Signal Based on Improved BPNN

A comparing network for the classification of steady-state visual evoked potential responses based on convolutional neural network

Automatic Classification of Epileption Activity in EEG Based on Wavelet Analysis and Teager Energy Operator

InceptionSSVEP: A Multi-Scale Convolutional Neural Network for Steady-State Visual Evoked Potential Classification

Circuit Breaker Fault Analysis Based on Wavelet Packet Time - Frequency Entropy and LM Algorithm to Optimize BP Neural Network

Feature Extraction of EEG for Imagery Left-Right Hands Movement