Publication: EOG Sinyalleri için Sınıflandırma Algoritmalarının Karşılaştırılması
Program
Authors
Advisor
Date
Language
Type
Publisher:
Journal Title
Journal ISSN
Volume Title
Abstract
Bu bildiride, belirli göz hareketleri sonucu elde edilen
elektrookülogram (EOG) sinyallerinin sınıflandırılması için
Destek Vektör Makinesi (DVM) ve Yapay Sinir Ağı (YSA)
yöntemleri karşılaştırılmaktadır. Göz hareketleri kullanarak
kontrol edilecek bir sistemin parçası olarak seçilen yöntemler,
doğruluk ve hız parametreleri bakımından karşılaştırılmıştır.
Alınan EOG sinyalleri, yatay (sağ ve sol), dikey (aşağı ve
yukarı) ve kırpma biçimindeki 5 farklı göz hareketinden
oluşmaktadır. 20 denekten, yatay ve dikey bileşenli iki EOG
kanalından alınan sinyal kümesinden, 3 elemanlı öznitelik
vektörleri hesaplanmıştır. Öznitelik vektörlerinin ilk iki
elemanı, kanallardaki tepe genlik değerinden, üçüncüsü ise bu
çalışmada özgün olarak önerilen parametre olan, aktif kanala
ait basıklık değerinden oluşmaktadır. Deneklerden elde edilen
öznitelik vektörlerinden rasgele 10 tanesi adı geçen
sınıflandırıcıların eğitimi, kalan 10 tanesi de başarımlarını
değerlendirmek için kullanılmıştır. Çevrimdışı olarak yapılan
testler sonucu her iki sınıflandırma yönteminde de % 100
başarı elde edilmiştir. Ortaya çıkan cevap süreleri de iki
yöntemin gerçek zamanlı olarak kullanıma uygun olduğunu
göstermektedir.
In this paper, we present a comparison Support Vector Machine (SVM) and Artificial Neural Network (ANN) for classification of electrooculogram (EOG) signals acquired under specific eye movements. These methods that are required for an eye controlled system are compared by means of their accuracy and response time. Acquired EOG signals consist of 5 different eye movements - being horizontal (right and left), vertical (up and down) and blink. EOG signal acquisition was achieved from 20 different subjects by using two EOG channels (vertical and horizontal) and 3 element feature vectors were extracted. The first two elements of the feature vectors are the peak amplitudes of two channels whereas the third element, being our proposed parameter, is the kurtosis value of the active channel. 10 of 20 randomly selected feature vectors were used for training of the classifiers whereas the rest was used for performance tests. Offline tests yield 100 % success rate for both of the classifiers. The response times of both methods make them suitable for real-time usage.
In this paper, we present a comparison Support Vector Machine (SVM) and Artificial Neural Network (ANN) for classification of electrooculogram (EOG) signals acquired under specific eye movements. These methods that are required for an eye controlled system are compared by means of their accuracy and response time. Acquired EOG signals consist of 5 different eye movements - being horizontal (right and left), vertical (up and down) and blink. EOG signal acquisition was achieved from 20 different subjects by using two EOG channels (vertical and horizontal) and 3 element feature vectors were extracted. The first two elements of the feature vectors are the peak amplitudes of two channels whereas the third element, being our proposed parameter, is the kurtosis value of the active channel. 10 of 20 randomly selected feature vectors were used for training of the classifiers whereas the rest was used for performance tests. Offline tests yield 100 % success rate for both of the classifiers. The response times of both methods make them suitable for real-time usage.
Description
Source:
Keywords:
Keywords
Electrooculography, Support vector machine classification, Barium, Artificial Neural Networks, Time Factors, Feature Extraction, Electro-Oculography, Medical Signal Processing, Neural Nets, Support Vector Machines, Classification Algorithms, EOG Signals, Electrooculogram Signals, Artificia lNeural Network, Support Vector Machine, SVM, ANN, Specific Eye Movements, Eye Controlled System, 3 Element Feature Vectors, Kurtosis Value, Offline Tests