Classification Electromyography Sensor for Hand Gesture Recognition with Neural Network using Correlation-Based Feature Selection
In this paper, we will present a research about hand gesture pattern recognition using electromyography sensors. Data collection process will take 100 samples in time domain data for each gesture, then transformed into frequency domain data and feature extraction to get 18 features. After getting the features, we will perform feature selection with Correlation-based Feature Selection (CFS) to reduce the total features and improve the accuracy of the classification algorithm used, which is the neural network. The analysis process will show the accuracy value of gestures used.
Index Terms- Electromyography Sensor, Feature Selection, Neural Network, Pattern Recognition.