A Smartphone-Based sEMG Signal Analysis System for Human Action Recognition

In lower-limb rehabilitation, human action recognition (HAR) technology can be introduced to analyze the surface electromyography (sEMG) signal generated by movements, which provide an objective and accurate evaluation of patient’s action. To balance long cycle required for rehabilitation inconvenient factors brought wearing sEMG devices, a portable acquisition device was developed that used under daily scenarios. Additionally, mobile application meet demand real-time monitoring analysis signals. This monitor data in real time has functions such as plotting, filtering, storage, capture recognition. build dataset model, six motions were (kick, toe off, heel off up, step back kick, full gait). The segment label combined training convolutional neural network (CNN) achieve high-precision performance actions (with maximum accuracy 97.96% all reaching over 97%). results show smartphone-based system proposed this paper reliable information clinical rehabilitation.