Scholarlink

In recent years, stab-resistant clothing has gained increasing importance as a key protective gear against terrorist attacks and violent incidents. However, the mechanical creases that form in puncture-resistant materials during production and use can significantly affect their protective performance. Traditional manual inspection methods are inefficient and prone to high errors, making them inadequate for modern protective material testing needs. This study simulates the formation of mechanical creases using a self-developed fatigue testing machine, combined with a custom-designed image acquisition system for high-precision crease image collection. A segmentation neural network is then utilized to recognize and annotate crease shapes and geometric features. Experimental results demonstrate that with increased bending time, both the peak puncture force and the number of penetration layers in AFRP materials significantly increase, indicating that prolonged bending can reconstruct the internal structure of the material, thereby enhancing its puncture resistance. The findings of this research provide valuable theoretical support for the design and performance evaluation of stab-resistant clothing, advancing the development of protective material testing technology.

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

Prediction of Crease Performance in Aramid Stab-Resistant Materials Based on Artificial Neural Networks and Surface Images

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Mengzhen Liu,
  • Siyu Liu,
  • Haotian Li,
  • Guangyan Huang,
  • Hong Zhang

摘要

In recent years, stab-resistant clothing has gained increasing importance as a key protective gear against terrorist attacks and violent incidents. However, the mechanical creases that form in puncture-resistant materials during production and use can significantly affect their protective performance. Traditional manual inspection methods are inefficient and prone to high errors, making them inadequate for modern protective material testing needs. This study simulates the formation of mechanical creases using a self-developed fatigue testing machine, combined with a custom-designed image acquisition system for high-precision crease image collection. A segmentation neural network is then utilized to recognize and annotate crease shapes and geometric features. Experimental results demonstrate that with increased bending time, both the peak puncture force and the number of penetration layers in AFRP materials significantly increase, indicating that prolonged bending can reconstruct the internal structure of the material, thereby enhancing its puncture resistance. The findings of this research provide valuable theoretical support for the design and performance evaluation of stab-resistant clothing, advancing the development of protective material testing technology.