الفهرس | يوجد فقط 14 صفحة متاحة للعرض العام |
المستخلص Brain-computer interface (BCl) is a new and promising area of research that is assumed to assist in solving a lot of problems especially for handicapped people. For example, the detection of left and right hand movements’ imagination can be used to control a wheelchair. Fortunately, the electrical brain activity due to left or right hand movements’ imagination is similar to the electrical brain activity of the real left or right hand movements. This activity can be picked up using scalp electroencephalography electrodes. In this work, new methods for the detection and classification of the imagination of the left and/or right hand movements are introduced. Each of These methods is divided into three stages. The first stage is preprocessing. In this stage, EEG signals are prepared for subsequent analysis. The second stage is features extraction. In this stage, the task-specific features from EEG signals are extracted using several mathematical techniques such as complex Morlet wavelet transform, autoregressive model, and principal component analysis. The last stage is classification. In this stage the extracted features of left or right hand movement imagination tasks are recognized using several mathematical techniques such as multi layer back- propagation neural network, support vector machine, and hidden Markov model. The experimental results explored that the proposed algorithms has revealed classi fication accuracy rates ranges from 96.11 % to 100%, that are superior to the classi fication accuracy rates of other techniques that are developed to solve this classi fication task. |