Deteksi Pergerakan Tangan dari Data EEG Menggunakan Fast Walsh–Hadamard Transform dan Artificial Neural Network

Palupi, Nurhamidah Tyas (2018) Deteksi Pergerakan Tangan dari Data EEG Menggunakan Fast Walsh–Hadamard Transform dan Artificial Neural Network. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
05111440000057-Udergraduate_theses.pdf - Published Version
Restricted to Repository staff only

Download (1MB) | Request a copy

Abstract

Brain computer interface merupakan sebuah teknologi yang memungkinkan individu dapat berkomunikasi dengan mesin tanpa melibatkan adanya pergerakan otot. Di dalam aplikasi BCI terbaru terdapat beberapa metode, diantaranya seperti electroencephalography (EEG), electrocorticography (EcoG), dan functional magnetic resonance imaging (fMRI). Namun dari ketiga metode tersebut EEG adalah metode yang lebih banyak digunakan, hal ini karena kemudahan dalam mendapatkan data serta menerapkannya. Dalam tugas akhir ini, penulis menerapkan sebuah metode untuk mendeteksi pergerakan tangan kanan atau kiri dari data sinyal EEG dengan menggunakan transformasi fitur Fast Walsh-Hadamart Transform (FWHT) dan metode klasifikasi Artificial Neural Network (ANN). Berdasarkan hasil pengujian yang dilakukan, menggunakan dua hidden layer dengan jumlah neuron [5 5] dan nilai learning rate 10 didapatkan akurasi tertinggi yaitu 73,79%. ============= Brain computer interface is a technology that allows individuals to communicate with machines without involving muscle movement. In the latest BCI applications there are several methods, such as electroencephalography (EEG), electrocorticography (EcoG), and functional magnetic resonance imaging (fMRI). But from the three methods above EEG is one of it which widely used, this because of the ease in obtaining and applying the data. In this final project the author implements a method for detecting left or right hand movements of EEG signals by using Fast Walsh-Hadamart Transform (FWHT) for feature extraction and Artificial Neural Network (ANN) for classification. Based on the results of tests performed, using two hidden layers with number of neurons [5 5] and learning rate 10 it can obtaine the highest accuracy that is 73,79%.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Brain computer interface (BCI), electroencephalography (EEG), Fast Walsh-Hadamart Transform, Artificial Neural Network (ANN).
Subjects: Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Q Science > QP Physiology > Q376.5 Electroencephalography (EEG)
T Technology > T Technology (General) > T57.5 Data Processing
Divisions: Faculty of Information and Communication Technology > Informatics > (S1) Undergraduate Theses
Depositing User: Nurhamidah Tyas Palupi
Date Deposited: 27 Jul 2018 08:03
Last Modified: 27 Jul 2018 08:03
URI: http://repository.its.ac.id/id/eprint/54096

Actions (login required)

View Item View Item