Myoelectricity data automatic labeling method

Abstract

The invention provides a myoelectricity data automatic labeling method. The myoelectricity data automatic labeling method comprises the following steps that S1, a synchronous collection mode of collecting myoelectricity data of hand actions by a myoelectricity device and collecting image data of the hand actions by LeapMotion is adopted; S2, gesture activity section extraction is performed on the LeapMotion gesture action data; S3, the LeapMotion gestures are clustered through a clustering method, and action labels are obtained. and S4, automatically labeling the myoelectricity gesture data is automatically according to the label obtained in the S3. According to the invention, the problem that a user is too boring and tired due to too long acquisition time of a certain action in gesture data acquisition can be solved. Meanwhile, the experiment platform for different identification tasks does not need to be additionally modified, so that the experiment platform modification time of an experimenter can be shortened, and the experiment progress is accelerated.

Description

technical field [0001] The invention belongs to the field of data processing, and in particular relates to a method for automatically labeling myoelectric data. Background technique [0002] With the emergence of a large number of smart devices and the development of artificial intelligence technology, the problem of human-computer interaction has gradually become a research hotspot. Among them, gesture recognition has the advantages of strong ease of use and practicality, and has attracted more and more people's attention. Gesture recognition can allow patients with physical disabilities to help them live a normal life by controlling prosthetics. Deaf-mute patients can communicate with healthy people through gesture recognition, and can also play interactive games through gestures to improve the fun of games. [0003] At present, gesture recognition methods are mainly divided into gesture recognition based on visual signals and gesture recognition based on electromyographi...

AI Technical Summary

Problems solved by technology

The traditional EMG signal acquisition experiment requires the subject to make corresponding actions according to the prompts of the experimental platform. Due to the hysteresis of the subject's response and the interference of the EMG signal noise, the data needs to be marked after the acquisition is completed to obtain a more accurate EMG signal. signal, which will increase the burden on the subjects and experimenters

The traditional EMG signal acquisition experiment needs to obtain different gesture data sets for different task requirements. The experimental platform needs to frequently modify the target gesture action, the number of repetitions of a certain gesture, and the sequence of gesture acquisition according to the requirements. poor repeatability

Images