Search results for: Log Gabor Wavelet

Authors: Somnath Dey, Debasis Samanta

Abstract:

Iris-based biometric authentication is gaining importance in recent times. Iris biometric processing however, is a complex process and computationally very expensive. In the overall processing of iris biometric in an iris-based biometric authentication system, feature processing is an important task. In feature processing, we extract iris features, which are ultimately used in matching. Since there is a large number of iris features and computational time increases as the number of features increases, it is therefore a challenge to develop an iris processing system with as few as possible number of features and at the same time without compromising the correctness. In this paper, we address this issue and present an approach to feature extraction and feature matching process. We apply Daubechies D4 wavelet with 4 levels to extract features from iris images. These features are encoded with 2 bits by quantizing into 4 quantization levels. With our proposed approach it is possible to represent an iris template with only 304 bits, whereas existing approaches require as many as 1024 bits. In addition, we assign different weights to different iris region to compare two iris templates which significantly increases the accuracy. Further, we match the iris template based on a weighted similarity measure. Experimental results on several iris databases substantiate the efficacy of our approach.

Keywords: Iris recognition, biometric, feature processing, patternrecognition, pattern matching.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2138

Authors: Sadi Vural, Hiromi Tomii, Hironori Yamauchi

Abstract:

With the advent of digital cinema and digital broadcasting, copyright protection of video data has been one of the most important issues. We present a novel method of watermarking for video image data based on the hardware and digital wavelet transform techniques and name it as “traceable watermarking" because the watermarked data is constructed before the transmission process and traced after it has been received by an authorized user. In our method, we embed the watermark to the lowest part of each image frame in decoded video by using a hardware LSI. Digital Cinema is an important application for traceable watermarking since digital cinema system makes use of watermarking technology during content encoding, encryption, transmission, decoding and all the intermediate process to be done in digital cinema systems. The watermark is embedded into the randomly selected movie frames using hash functions. Embedded watermark information can be extracted from the decoded video data. For that, there is no need to access original movie data. Our experimental results show that proposed traceable watermarking method for digital cinema system is much better than the convenient watermarking techniques in terms of robustness, image quality, speed, simplicity and robust structure.

Keywords: Decoder, Digital content, JPEG2000 Frame, System-On-Chip, traceable watermark, Hash Function, CRC-32.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1646

Authors: Samira Vafaye Eslahi

Abstract:

BCI (Brain Computer Interface) is a communication machine that translates brain massages to computer commands. These machines with the help of computer programs can recognize the tasks that are imagined. Feature extraction is an important stage of the process in EEG classification that can effect in accuracy and the computation time of processing the signals. In this study we process the signal in three steps of active segment selection, fractal feature extraction, and classification. One of the great challenges in BCI applications is to improve classification accuracy and computation time together. In this paper, we have used student’s 2D sample t-statistics on continuous wavelet transforms for active segment selection to reduce the computation time. In the next level, the features are extracted from some famous fractal dimension estimation of the signal. These fractal features are Katz and Higuchi. In the classification stage we used ANFIS (Adaptive Neuro-Fuzzy Inference System) classifier, FKNN (Fuzzy K-Nearest Neighbors), LDA (Linear Discriminate Analysis), and SVM (Support Vector Machines). We resulted that active segment selection method would reduce the computation time and Fractal dimension features with ANFIS analysis on selected active segments is the best among investigated methods in EEG classification.

Keywords: EEG, Student’s t- statistics, BCI, Fractal Features, ANFIS, FKNN.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2120

Authors: Thomas Bryan, Veton Kepuska, Ivica Kostanic

Abstract:

Speaker recognition is performed in high Additive White Gaussian Noise (AWGN) environments using principals of Computational Auditory Scene Analysis (CASA). CASA methods often classify sounds from images in the time-frequency (T-F) plane using spectrograms or cochleargrams as the image. In this paper atomic decomposition implemented by matching pursuit performs a transform from time series speech signals to the T-F plane. The atomic decomposition creates a sparsely populated T-F vector in “weight space” where each populated T-F position contains an amplitude weight. The weight space vector along with the atomic dictionary represents a denoised, compressed version of the original signal. The arraignment or of the atomic indices in the T-F vector are used for classification. Unsupervised feature learning implemented by a sparse autoencoder learns a single dictionary of basis features from a collection of envelope samples from all speakers. The approach is demonstrated using pairs of speakers from the TIMIT data set. Pairs of speakers are selected randomly from a single district. Each speak has 10 sentences. Two are used for training and 8 for testing. Atomic index probabilities are created for each training sentence and also for each test sentence. Classification is performed by finding the lowest Euclidean distance between then probabilities from the training sentences and the test sentences. Training is done at a 30dB Signal-to-Noise Ratio (SNR). Testing is performed at SNR’s of 0 dB, 5 dB, 10 dB and 30dB. The algorithm has a baseline classification accuracy of ~93% averaged over 10 pairs of speakers from the TIMIT data set. The baseline accuracy is attributable to short sequences of training and test data as well as the overall simplicity of the classification algorithm. The accuracy is not affected by AWGN and produces ~93% accuracy at 0dB SNR.

Keywords: Time-frequency plane, atomic decomposition, envelope sampling, Gabor atoms, matching pursuit, sparse dictionary learning, sparse autoencoder.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1570

Authors: Cameron Hamilton, Walter Potter, Gerrit Hoogenboom, Ronald McClendon, Will Hobbs

Abstract:

A model was constructed to predict the amount of solar radiation that will make contact with the surface of the earth in a given location an hour into the future. This project was supported by the Southern Company to determine at what specific times during a given day of the year solar panels could be relied upon to produce energy in sufficient quantities. Due to their ability as universal function approximators, an artificial neural network was used to estimate the nonlinear pattern of solar radiation, which utilized measurements of weather conditions collected at the Griffin, Georgia weather station as inputs. A number of network configurations and training strategies were utilized, though a multilayer perceptron with a variety of hidden nodes trained with the resilient propagation algorithm consistently yielded the most accurate predictions. In addition, a modeled direct normal irradiance field and adjacent weather station data were used to bolster prediction accuracy. In later trials, the solar radiation field was preprocessed with a discrete wavelet transform with the aim of removing noise from the measurements. The current model provides predictions of solar radiation with a mean square error of 0.0042, though ongoing efforts are being made to further improve the model’s accuracy.

Keywords: Artificial Neural Networks, Resilient Propagation, Solar Radiation, Time Series Forecasting.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2761

Authors: L. Bennaceur Farah, I. R. Farah, R. Bennaceur, Z. Belhadj, M. R. Boussema

Abstract:

The overall objective of this paper is to retrieve soil surfaces parameters namely, roughness and soil moisture related to the dielectric constant by inverting the radar backscattered signal from natural soil surfaces. Because the classical description of roughness using statistical parameters like the correlation length doesn't lead to satisfactory results to predict radar backscattering, we used a multi-scale roughness description using the wavelet transform and the Mallat algorithm. In this description, the surface is considered as a superposition of a finite number of one-dimensional Gaussian processes each having a spatial scale. A second step in this study consisted in adapting a direct model simulating radar backscattering namely the small perturbation model to this multi-scale surface description. We investigated the impact of this description on radar backscattering through a sensitivity analysis of backscattering coefficient to the multi-scale roughness parameters. To perform the inversion of the small perturbation multi-scale scattering model (MLS SPM) we used a multi-layer neural network architecture trained by backpropagation learning rule. The inversion leads to satisfactory results with a relative uncertainty of 8%.

Keywords: Remote sensing, rough surfaces, inverse problems, SAR, radar scattering, Neural networks and Fractals.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1594

Authors: Maria C. Proença, Marco Aniceto, Pedro N. Santos, José C. Freitas

Abstract:

A methodology based on wavelets is proposed for the automatic location and delimitation of defects in limestone plates. Natural defects include dark colored spots, crystal zones trapped in the stone, areas of abnormal contrast colors, cracks or fracture lines, and fossil patterns. Although some of these may or may not be considered as defects according to the intended use of the plate, the goal is to pair each stone with a map of defects that can be overlaid on a computer display. These layers of defects constitute a database that will allow the preliminary selection of matching tiles of a particular variety, with specific dimensions, for a requirement of N square meters, to be done on a desktop computer rather than by a two-hour search in the storage park, with human operators manipulating stone plates as large as 3 m x 2 m, weighing about one ton. Accident risks and work times are reduced, with a consequent increase in productivity. The base for the algorithm is wavelet decomposition executed in two instances of the original image, to detect both hypotheses – dark and clear defects. The existence and/or size of these defects are the gauge to classify the quality grade of the stone products. The tuning of parameters that are possible in the framework of the wavelets corresponds to different levels of accuracy in the drawing of the contours and selection of the defects size, which allows for the use of the map of defects to cut a selected stone into tiles with minimum waste, according the dimension of defects allowed.

Keywords: Automatic detection, wavelets, defects, fracture lines.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1166

Authors: Xiang Jianhong, Wang Cong, Wang Linyu

Abstract:

With the rise of medical IoT technologies, Wireless body area networks (WBANs) can collect fetal electrocardiogram (FECG) signals to support telemedicine analysis. The compressed sensing (CS)-based WBANs system can avoid the sampling of a large amount of redundant information and reduce the complexity and computing time of data processing, but the existing algorithms have poor signal compression and reconstruction performance. In this paper, a Joint block multi-orthogonal least squares (JBMOLS) algorithm is proposed. We apply the FECG signal to the Joint block sparse model (JBSM), and a comparative study of sparse transformation and measurement matrices is carried out. A FECG signal compression transmission mode based on Rbio5.5 wavelet, Bernoulli measurement matrix, and JBMOLS algorithm is proposed to improve the compression and reconstruction performance of FECG signal by CS-based WBANs. Experimental results show that the compression ratio (CR) required for accurate reconstruction of this transmission mode is increased by nearly 10%, and the runtime is saved by about 30%.

Keywords: telemedicine, fetal electrocardiogram, compressed sensing, joint sparse reconstruction, block sparse signal

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 510

Authors: Sule Yucelbas, Gulay Tezel, Cuneyt Yucelbas, Seral Ozsen

Abstract:

In this study, it was tried to identify some heart rhythm disorders by electrocardiography (ECG) data that is taken from MIT-BIH arrhythmia database by subtracting the required features, presenting to artificial neural networks (ANN), artificial immune systems (AIS), artificial neural network based on artificial immune system (AIS-ANN) and particle swarm optimization based artificial neural network (PSO-NN) classifier systems. The main purpose of this study is to evaluate the performance of hybrid AIS-ANN and PSO-ANN classifiers with regard to the ANN and AIS. For this purpose, the normal sinus rhythm (NSR), atrial premature contraction (APC), sinus arrhythmia (SA), ventricular trigeminy (VTI), ventricular tachycardia (VTK) and atrial fibrillation (AF) data for each of the RR intervals were found. Then these data in the form of pairs (NSR-APC, NSR-SA, NSR-VTI, NSR-VTK and NSR-AF) is created by combining discrete wavelet transform which is applied to each of these two groups of data and two different data sets with 9 and 27 features were obtained from each of them after data reduction. Afterwards, the data randomly was firstly mixed within themselves, and then 4-fold cross validation method was applied to create the training and testing data. The training and testing accuracy rates and training time are compared with each other.

As a result, performances of the hybrid classification systems, AIS-ANN and PSO-ANN were seen to be close to the performance of the ANN system. Also, the results of the hybrid systems were much better than AIS, too. However, ANN had much shorter period of training time than other systems. In terms of training times, ANN was followed by PSO-ANN, AIS-ANN and AIS systems respectively. Also, the features that extracted from the data affected the classification results significantly.

Keywords: AIS, ANN, ECG, hybrid classifiers, PSO.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1916

Authors: Nassib Abdallah, Pierre Chauvet, Abd El Salam Hajjar, Bassam Daya

Abstract:

In this paper, we propose an optimized brain computer interface (BCI) system for unspoken speech recognition, based on the fact that the constructions of unspoken words rely strongly on the Wernicke area, situated in the temporal lobe. Our BCI system has four modules: (i) the EEG Acquisition module based on a non-invasive headset with 14 electrodes; (ii) the Preprocessing module to remove noise and artifacts, using the Common Average Reference method; (iii) the Features Extraction module, using Wavelet Packet Transform (WPT); (iv) the Classification module based on a one-hidden layer artificial neural network. The present study consists of comparing the recognition accuracy of 5 Arabic words, when using all the headset electrodes or only the 4 electrodes situated near the Wernicke area, as well as the selection effect of the subbands produced by the WPT module. After applying the articial neural network on the produced database, we obtain, on the test dataset, an accuracy of 83.4% with all the electrodes and all the subbands of 8 levels of the WPT decomposition. However, by using only the 4 electrodes near Wernicke Area and the 6 middle subbands of the WPT, we obtain a high reduction of the dataset size, equal to approximately 19% of the total dataset, with 67.5% of accuracy rate. This reduction appears particularly important to improve the design of a low cost and simple to use BCI, trained for several words.

Keywords: Brain-computer interface, speech recognition, electroencephalography EEG, Wernicke area, artificial neural network.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 918

Authors: Christopher Paterson, Richard Curry, Alan Purvis, Simon Johnson

Abstract:

Emerging Bio-engineering fields such as Brain Computer Interfaces, neuroprothesis devices and modeling and simulation of neural networks have led to increased research activity in algorithms for the detection, isolation and classification of Action Potentials (AP) from noisy data trains. Current techniques in the field of 'unsupervised no-prior knowledge' biosignal processing include energy operators, wavelet detection and adaptive thresholding. These tend to bias towards larger AP waveforms, AP may be missed due to deviations in spike shape and frequency and correlated noise spectrums can cause false detection. Also, such algorithms tend to suffer from large computational expense. A new signal detection technique based upon the ideas of phasespace diagrams and trajectories is proposed based upon the use of a delayed copy of the AP to highlight discontinuities relative to background noise. This idea has been used to create algorithms that are computationally inexpensive and address the above problems. Distinct AP have been picked out and manually classified from real physiological data recorded from a cockroach. To facilitate testing of the new technique, an Auto Regressive Moving Average (ARMA) noise model has been constructed bases upon background noise of the recordings. Along with the AP classification means this model enables generation of realistic neuronal data sets at arbitrary signal to noise ratio (SNR).

Keywords: Action potential detection, Low SNR, Phase spacediagrams/trajectories, Unsupervised/no-prior knowledge.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1642

Authors: P. Sophon, M. Kruachanta, S. Chaisri, G. Leaungvongpaisan, P. Wongpornchai

Abstract:

The thick bed hydrocarbon reservoirs are primarily interested because of the more prolific production. When the amount of petroleum in the thick bed starts decreasing, the thin bed reservoirs are the alternative targets to maintain the reserves. The conventional interpretation of seismic data cannot delineate the thin bed having thickness less than the vertical seismic resolution. Therefore, spectral decomposition and instantaneous seismic attributes were used to delineate the thin bed in this study. Short Window Discrete Fourier Transform (SWDFT) spectral decomposition and instantaneous frequency attributes were used to reveal the thin bed reservoir, while Continuous Wavelet Transform (CWT) spectral decomposition and envelope (instantaneous amplitude) attributes were used to indicate hydrocarbon bearing zone. The study area is located in the Pohokura Field, Taranaki Basin, New Zealand. The thin bed target is the uppermost part of Mangahewa Formation, the most productive in the gas-condensate production in the Pohokura Field. According to the time-frequency analysis, SWDFT spectral decomposition can reveal the thin bed using a 72 Hz SWDFT isofrequency section and map, and that is confirmed by the instantaneous frequency attribute. The envelope attribute showing the high anomaly indicates the hydrocarbon accumulation area at the thin bed target. Moreover, the CWT spectral decomposition shows the low-frequency shadow zone and abnormal seismic attenuation in the higher isofrequencies below the thin bed confirms that the thin bed can be a prospective hydrocarbon zone.

Keywords: Hydrocarbon indication, instantaneous seismic attribute, spectral decomposition, thin bed delineation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 640

Authors: Zsolt János Viharos, Krisztián Balázs Kis, Imre Paniti, Gábor Belső, Péter Németh, János Farkas

Abstract:

The paper presents the results and industrial applications in the production setup period estimation based on industrial data inherited from the field of polymer cutting. The literature of polymer cutting is very limited considering the number of publications. The first polymer cutting machine is known since the second half of the 20th century; however, the production of polymer parts with this kind of technology is still a challenging research topic. The products of the applying industrial partner must met high technical requirements, as they are used in medical, measurement instrumentation and painting industry branches. Typically, 20% of these parts are new work, which means every five years almost the entire product portfolio is replaced in their low series manufacturing environment. Consequently, it requires a flexible production system, where the estimation of the frequent setup periods' lengths is one of the key success factors. In the investigation, several (input) parameters have been studied and grouped to create an adequate training information set for an artificial neural network as a base for the estimation of the individual setup periods. In the first group, product information is collected such as the product name and number of items. The second group contains material data like material type and colour. In the third group, surface quality and tolerance information are collected including the finest surface and tightest (or narrowest) tolerance. The fourth group contains the setup data like machine type and work shift. One source of these parameters is the Manufacturing Execution System (MES) but some data were also collected from Computer Aided Design (CAD) drawings. The number of the applied tools is one of the key factors on which the industrial partners’ estimations were based previously. The artificial neural network model was trained on several thousands of real industrial data. The mean estimation accuracy of the setup periods' lengths was improved by 30%, and in the same time the deviation of the prognosis was also improved by 50%. Furthermore, an investigation on the mentioned parameter groups considering the manufacturing order was also researched. The paper also highlights the manufacturing introduction experiences and further improvements of the proposed methods, both on the shop floor and on the quotation preparation fields. Every week more than 100 real industrial setup events are given and the related data are collected.

Keywords: Artificial neural network, low series manufacturing, polymer cutting, setup period estimation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 975

Authors: Wernhuar Tarng, Yuan-Yuan Chen, Chien-Lung Li, Kun-Rong Hsie, Mingteh Chen

Abstract:

An emotional speech recognition system for the applications on smart phones was proposed in this study to combine with 3G mobile communications and social networks to provide users and their groups with more interaction and care. This study developed a mechanism using the support vector machines (SVM) to recognize the emotions of speech such as happiness, anger, sadness and normal. The mechanism uses a hierarchical classifier to adjust the weights of acoustic features and divides various parameters into the categories of energy and frequency for training. In this study, 28 commonly used acoustic features including pitch and volume were proposed for training. In addition, a time-frequency parameter obtained by continuous wavelet transforms was also used to identify the accent and intonation in a sentence during the recognition process. The Berlin Database of Emotional Speech was used by dividing the speech into male and female data sets for training. According to the experimental results, the accuracies of male and female test sets were increased by 4.6% and 5.2% respectively after using the time-frequency parameter for classifying happy and angry emotions. For the classification of all emotions, the average accuracy, including male and female data, was 63.5% for the test set and 90.9% for the whole data set.

Keywords: Smart phones, emotional speech recognition, socialnetworks, support vector machines, time-frequency parameter, Mel-scale frequency cepstral coefficients (MFCC).

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1841

Authors: B. Elshafei, X. Mao

Abstract:

The demand for renewable energy is significantly increasing, major investments are being supplied to the wind power generation industry as a leading source of clean energy. The wind energy sector is entirely dependable and driven by the prediction of wind speed, which by the nature of wind is very stochastic and widely random. This s0tudy employs deep multi-fidelity Gaussian process regression, used to predict wind speeds for medium term time horizons. Data of the RUNE experiment in the west coast of Denmark were provided by the Technical University of Denmark, which represent the wind speed across the study area from the period between December 2015 and March 2016. The study aims to investigate the effect of pre-processing the data by denoising the signal using empirical wavelet transform (EWT) and engaging the vector components of wind speed to increase the number of input data layers for data fusion using deep multi-fidelity Gaussian process regression (GPR). The outcomes were compared using root mean square error (RMSE) and the results demonstrated a significant increase in the accuracy of predictions which demonstrated that using vector components of the wind speed as additional predictors exhibits more accurate predictions than strategies that ignore them, reflecting the importance of the inclusion of all sub data and pre-processing signals for wind speed forecasting models.

Keywords: Data fusion, Gaussian process regression, signal denoise, temporal extrapolation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 501

Authors: Arvind Srivastav, Tarun Kanti Bhattacharyya

Abstract:

The north-eastern, Himalayan, and Eastern Ghats Belt of India comprise of earthquake-prone, remote, and hilly terrains. Earthquakes have caused enormous damages in these regions in the past. A wireless sensor network based earthquake early warning system (EEWS) is being developed to mitigate the damages caused by earthquakes. It consists of sensor nodes, distributed over the region, that perform majority voting of the output of the seismic sensors in the vicinity, and relay a message to a base station to alert the residents when an earthquake is detected. At the heart of the EEWS is a low-power two-stage seismic sensor that continuously tracks seismic events from incoming three-axis accelerometer signal at the first-stage, and, in the presence of a seismic event, triggers the second-stage P-wave detector that detects the onset of P-wave in an earthquake event. The parameters of the P-wave detector have been optimized for minimizing detection time and maximizing the accuracy of detection.Working of the sensor scheme has been verified with seven earthquakes data retrieved from IRIS. In all test cases, the scheme detected the onset of P-wave accurately. Also, it has been established that the P-wave onset detection time reduces linearly with the sampling rate. It has been verified with test data; the detection time for data sampled at 10Hz was around 2 seconds which reduced to 0.3 second for the data sampled at 100Hz.

Keywords: Earthquake early warning system, EEWS, STA/LTA, polarization, wavelet, event detector, P-wave detector.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 781

Authors: Justinian Anatory

Abstract:

power-line networks are promise infrastructure for broadband services provision to end users. However, the network performance is affected by stochastic channel changing which is due to load impedances, number of branches and branched line lengths. It has been proposed that multi-carrier modulations techniques such as orthogonal frequency division multiplexing (OFDM), Multi-Carrier Spread Spectrum (MC-SS), wavelet OFDM can be used in such environment. This paper investigates the performance of different indoor topologies of power-line networks that uses MC-SS modulation scheme.It is observed that when a branch is added in the link between sending and receiving end of an indoor channel an average of 2.5dB power loss is found. In additional, when the branch is added at a node an average of 1dB power loss is found. Additionally when the terminal impedances of the branch change from line characteristic impedance to impedance either higher or lower values the channel performances were tremendously improved. For example changing terminal load from characteristic impedance (85 .) to 5 . the signal to noise ratio (SNR) required to attain the same performances were decreased from 37dB to 24dB respectively. Also, changing the terminal load from channel characteristic impedance (85 .) to very higher impedance (1600 .) the SNR required to maintain the same performances were decreased from 37dB to 23dB. The result concludes that MC-SS performs better compared with OFDM techniques in all aspects and especially when the channel is terminated in either higher or lower impedances.

Keywords: Communication channel model; Broadband Powerlinecommunication; Branched network; OFDM; Delay Spread, MCSS;impulsive noise; load impedance

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1606

Authors: M. R. Homaeinezhad, E. Tavakkoli, M. Habibi, S. A. Atyabi, A. Ghaffari

Abstract:

This paper describes a new supervised fusion (hybrid) electrocardiogram (ECG) classification solution consisting of a new QRS complex geometrical feature extraction as well as a new version of the learning vector quantization (LVQ) classification algorithm aimed for overcoming the stability-plasticity dilemma. Toward this objective, after detection and delineation of the major events of ECG signal via an appropriate algorithm, each QRS region and also its corresponding discrete wavelet transform (DWT) are supposed as virtual images and each of them is divided into eight polar sectors. Then, the curve length of each excerpted segment is calculated and is used as the element of the feature space. To increase the robustness of the proposed classification algorithm versus noise, artifacts and arrhythmic outliers, a fusion structure consisting of five different classifiers namely as Support Vector Machine (SVM), Modified Learning Vector Quantization (MLVQ) and three Multi Layer Perceptron-Back Propagation (MLP–BP) neural networks with different topologies were designed and implemented. The new proposed algorithm was applied to all 48 MIT–BIH Arrhythmia Database records (within–record analysis) and the discrimination power of the classifier in isolation of different beat types of each record was assessed and as the result, the average accuracy value Acc=98.51% was obtained. Also, the proposed method was applied to 6 number of arrhythmias (Normal, LBBB, RBBB, PVC, APB, PB) belonging to 20 different records of the aforementioned database (between– record analysis) and the average value of Acc=95.6% was achieved. To evaluate performance quality of the new proposed hybrid learning machine, the obtained results were compared with similar peer– reviewed studies in this area.

Keywords: Feature Extraction, Curve Length Method, SupportVector Machine, Learning Vector Quantization, Multi Layer Perceptron, Fusion (Hybrid) Classification, Arrhythmia Classification, Supervised Learning Machine.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2225