LSTM Network Classification of Dexterous Individual Finger Movements

Electrical activity is generated in the forearm muscles during muscular contractions that control dexterous movements of a human finger and thumb. Using this electrical as an input to train neural network for purposes classifying not straightforward. Low cost wearable sensors i.e., Myo Gesture armband (www.bynorth.com), generally have lower sampling rate when compared with medical grade EMG detection systems e.g., 200 Hz vs 2000 Hz. such coupled amplitude by individual makes it difficult achieve high classification accuracy. challenging distinguish between large quantities subtle using single network. This research uses two networks which enables reduction number each are being classified; turn improving classification. achieved developing training LSTM focus on extension flexion signals fingers separate trained thumb movement signal data. By following method, increased 90 100%.