Training data of Human Motion Recognition via Wearable plastic Fiber Sensing System

Human motion recognition based on SVM:

First of all, feature extraction and feature normalization of the preprocessed signals are required, then principal component analysis (PCA) is used for feature dimension reduction, and finally, SVM algorithm is used to recognize and classify human motion.

Human movement recognition based on MobileNet-v2 network and transfer learning:

Firstly, the motion signal of one-dimensional time series should be converted into two-dimensional image by GAF algorithm, and then the human movement can be recognized and classified by MobileNet-v2 network and transfer learning.