Figure 5 - uploaded by Kjersti Brynestad
Content may be subject to copyright.
Source publication
This work investigates possible approaches to developing a motor imagery-based Brain-Computer Interface (BCI) that can be used to actuate a drone. By using Electroencephalography (EEG) to record brain activity during mental execution of the movement of limbs, different imagined movements can be classified and act as commands for the drone. The main...
Similar publications
Brain-Computer Interfaces (BCI) based on Motor Imagery (MI) extract commands in real time and can be used to control a cursor, wheelchair, robot or prosthesis by performing only mental imaging tasks, such as imagining a movement of the right hand while the corresponding brain activity is measured and processed by the system. Because MI-based BCI of...
Motor imagery (MI)-based brain-computer interfaces have gained much attention in the last few years. They provide the ability to control external devices, such as prosthetic arms and wheelchairs, by using brain activities. Several researchers have reported the inter-communication of multiple brain regions during motor tasks, thus making it difficul...
In the field of Brain Computer Interface (BCI), applications in real life like emotion recognition from recorded electrical activity from brain have become famous topic of research nowadays. Learning successful representations of consistent performances from electroencephalogram (EEG) signals is one of the difficulties in recognition tasks. This re...
![Figure 1. International 10/20 EEG system cap montage [38].](publication/371023154/figure/fig1/AS:11431281162135617@1685136372366/International-10-20-EEG-system-cap-montage-38_Q320.jpg)
![Figure 2. Paradigm of the experiment trial of ULM dataset [39].](publication/371023154/figure/fig2/AS:11431281162110096@1685136372620/Paradigm-of-the-experiment-trial-of-ULM-dataset-39_Q320.jpg)
![Figure 3. Paradigm of the experiment trial of the KGU dataset [38]. The...](publication/371023154/figure/fig3/AS:11431281162128562@1685136372685/Paradigm-of-the-experiment-trial-of-the-KGU-dataset-38-The-top-row-of-the-image_Q320.jpg)
Motor imagery (MI) is a technique of imagining the performance of a motor task without actually using the muscles. When employed in a brain–computer interface (BCI) supported by electroencephalographic (EEG) sensors, it can be used as a successful method of human–computer interaction. In this paper, the performance of six different classifiers, nam...
Most brain–computer interface (BCI) systems operate on the basis of electroencephalography (EEG) due to their straightforwardness of applications and high temporal resolution of brain signals. One of the most helpful tools in BCI systems is the steady-state visual evoked potential, which is derived from the EEG data. This work evaluates a number of...
Citations
... In this project the wavelet Biorthogonal 2.8 (bior2.8) is used as the mother wavelet. This is because it has shown satisfying results in previous research on DWT used on EEG [38]. It also has symmetric and biorthogonal properties, which is useful in EEG, as shown by previous studies [39]. ...
This project tests out a possible Brain-Computer Interface (BCI) design which can be used for real-time drone control. The tested design utilizes motor imagery (MI) where the user imagines limb movement, without occurrence of physical movement. The MI is collected using electroencephalog-raphy (EEG). The main motivation and challenge of this project is to create a reliable BCI system which will help the drone serve as an addition to human functions for the physically disabled. The proposed design will be tested on two datasets; dataset NTNU and dataset IV2b. Both datasets contains motor imagery tasks from two classes for classification. Six different classification methods will be tested. The signal will be decomposed by either Empirical Mode Decomposition (EMD), Discrete Wavelet Transform (DWT) or Frequency Band Extraction (FBE). Features will then be extracted from the decomposed signal. These features will then be used as input in either Random Forest (RF) or Gradient Boosting (GB), for classification. No final conclusion was made for which classification method was the best. Subject variability, dataset quality and trade off between run-time and accuracy, were some factors which prevented this. Therefore , further analysis will be needed to make a definite conclusion. Nevertheless, the results indicated that these methods could be used to separate 2-class MI. However, this was only valid for dataset IV2b, which at best got an accuracy of 78% for all subjects, and 97% for a specific subject. The NTNU dataset gave unsatisfying results which were comparable to a statistical random guess at 50%. This occured due to some subjects lacking the ability to evoke the expected MI-based brain responses. Nonetheless , as the method was satisfactory for the commonly used IV2b dataset, these results are viable for MI-based BCI.
