Scholarlink

This paper introduces an improved neural network that integrates residual network and attention mechanisms, achieving a significant improvement of 99.50% accuracy in mechanical part classification tasks. The proposed network enhance accuracy, precision, recall, and F1-score compared to traditional CNNs and other advanced networks in handling complex backgrounds and multi-angle part images. This advancement shows potential in smart manufacturing, offering greater automation, minimizing human intervention.

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

Application of an Improved Residual Attention Neural Network in Mechanical Part Classification

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Tianrui Zhang,
  • Aizeng Wang

摘要

This paper introduces an improved neural network that integrates residual network and attention mechanisms, achieving a significant improvement of 99.50% accuracy in mechanical part classification tasks. The proposed network enhance accuracy, precision, recall, and F1-score compared to traditional CNNs and other advanced networks in handling complex backgrounds and multi-angle part images. This advancement shows potential in smart manufacturing, offering greater automation, minimizing human intervention.