应用于机器学习的手势肌电信号识别训练数据集

应用于机器学习的手势肌电信号识别训练数据集通过柔性肌电传感器采集人体上臂肌肉8个区域做出9种操作手势时的肌电信号。对采集到的肌电信号数据，分别进行降噪预处理、数据整合、过滤高噪声和工频干扰后，形成手势肌电信号的训练数据集和测试集。数据集输入卷积神经网络（CNN）训练得到对应的手势肌电信号标签，用以支撑机器识别手势肌电信号操作指令的功能。

A training dataset for gesture electromyography (EMG) signal recognition in machine learning applications is constructed by collecting EMG signals from eight regions of the human upper arm muscles when nine operational gestures are performed, using flexible EMG sensors. The collected EMG data are sequentially processed via noise reduction preprocessing, data integration, high-noise filtering and power line interference removal, and then split into the gesture EMG signal training dataset and test set. When fed into a Convolutional Neural Network (CNN) for training, this dataset generates corresponding labels for gesture EMG signals, which supports the machine's capability to recognize operation commands based on gesture EMG signals.

搜集汇总

背景与挑战

背景概述

该数据集是一个手势肌电信号识别训练数据集，包含8个区域9种手势的肌电信号，经过预处理后用于CNN模型训练，支持机器识别手势操作。数据集格式为CSV，数据量为3000条。