Scholarlink

Multispectral imaging systems used in the area of advanced material analysis are becoming efficient means for information search and object identification. This technology, especially using the Visible and Near-Infrared (VNIR) unit from 300 to 900 nm, will improve the ability to distinguish between different metals with a high degree of accuracy. This work seeks to classify metals like copper, aluminum, mild steel, stainless steel, and wood through the distinguishable spectral characteristics. Using the advanced techniques of machine learning, we have designed a highly effective classification system that greatly enhances metal identification. This paper presents the findings that proves the effectiveness of using VNIR spectroscopy with machine learning algorithms.

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

Metal Classification Using Visible Near-Infrared, ML and Spectroscopy

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Karma Shah,
  • Tanvi Bhat,
  • Rishabh Soni,
  • Manav Upadhyay,
  • Hardikkumar S. Jayswal,
  • Jitendra P. Chaudhari,
  • Nilesh Dubey,
  • Dipika Damodar,
  • Shital Sharma,
  • Axat Patel

摘要

Multispectral imaging systems used in the area of advanced material analysis are becoming efficient means for information search and object identification. This technology, especially using the Visible and Near-Infrared (VNIR) unit from 300 to 900 nm, will improve the ability to distinguish between different metals with a high degree of accuracy. This work seeks to classify metals like copper, aluminum, mild steel, stainless steel, and wood through the distinguishable spectral characteristics. Using the advanced techniques of machine learning, we have designed a highly effective classification system that greatly enhances metal identification. This paper presents the findings that proves the effectiveness of using VNIR spectroscopy with machine learning algorithms.