A Multiple Gait phase Recognition Method using Boosted Classifiers and Classification Matrix based on Surface EMG Signals

摘要

invention does not use a physical sensor surface electromyography (sEMG, surface electromyography) signals and only use two classifiers (Heel LDA, Toe LDA) used in the stance phase 3 (Heel strike , Mid stance, Toe off) and oil as each relates to multiple walking steps recognition method using surface EMG signals based boost classifiers and classification matrix to classify the first stage (swing), (a) a sample signal from the EMG signals of multiple channels receiving, generating a classifier for classifying the EMG signal; (B) Establish a configuration that includes the step of recognizing the walking motion by the classifier. And by the way, it can be classified by the walking step without the use of physical sensor using only a surface electromyogram (sEMG) signal. ;