Scholarlink

This study compares a convolutional neural network (CNN) and a multilayer perceptron (MLP) for the classification of the biceps brachii, gastrocnemius, and extensor digitorum muscles using surface electromyography (sEMG) signals. Using spectrograms and GoogLeNet, the CNN achieved an accuracy of 92.59%, while employing Neural Net Pattern Recognition and 6 features selected by the Kruskal-Wallis Method (KWM), the MLP achieved an accuracy of 71.1%. The superior performance of the CNN is attributed to its ability to automatically extract important features and handle complex information.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Analysis and Classification of Body Muscles during Contraction Using EMG Signals and Neural Networks

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Portos Juárez Francisco Josué,
  • Lima Zempoaltecatl Adrian,
  • Félix García Edgar Antonio,
  • Hurtado Pérez Andrés Emilio

摘要

This study compares a convolutional neural network (CNN) and a multilayer perceptron (MLP) for the classification of the biceps brachii, gastrocnemius, and extensor digitorum muscles using surface electromyography (sEMG) signals. Using spectrograms and GoogLeNet, the CNN achieved an accuracy of 92.59%, while employing Neural Net Pattern Recognition and 6 features selected by the Kruskal-Wallis Method (KWM), the MLP achieved an accuracy of 71.1%. The superior performance of the CNN is attributed to its ability to automatically extract important features and handle complex information.