A Deep Learning based Optimization Model for Based Computer Interface of Wheelchair Directional Control

An efficient recognition model is highly recommended while trying to analyze brain signal pattern for Motor Imagery (MI) signal. Therefore, this study aims develop an optimized based on a deep learning approach using Multi-Layer Perceptron (MLP) in order help large community of disability people by allowing them control the wheelchair their MI Brain In paper, dataset used which belong BCI Competition IV/2b and consists two parts, each contains 160 trails single subject. To preprocess signal, Butterworth band pass filter remove unwanted (Alpha Beta) remain then followed feature extraction technique Discrete Wavelet Transform (DWT). After that, classifier training parameters utilized optimize performance proposed system through grid search optimization improve distinguishing between directional commands. Cross-validations with ten groups were adopted boost modeling accuracy all subjects (1440 trials) (160 trails). The results showed that efficiency MLP increased 3% over compared non-optimized model. It can be concluded deployed daily activities.