Scholarlink

Muon Scattering Tomography (MST) is an effective technique for identifying special nuclear materials (SNM). Images of SNM and other materials in cargo can be produced on the basis of scattering suffered by cosmic mons while passing through the objects. In the present work, few image processing protocols are proposed with the help of simulated images for a prototype MST setup which is currently under construction. The images were produced from analyzing scattering angles within the target material using the Point of Closest Approach (PoCA) algorithm and further processed with a Pattern Recognition Method (PRM). Following this, Deep Convolutional Neural Network (DCNN) and Semantic Image Segmentation (SIS) models were employed to classify different materials.

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

Material Identification and Multiclass Classification of High-Z Materials in Cosmic Muon Imaging Using PRM and Deep CNN SIS Model

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Saikat Ghosh,
  • Sreeja Singh,
  • Shubhabrata Dutta,
  • Subhendu Das,
  • Nayana Majumdar,
  • Supratik Mukhopadhyay

摘要

Muon Scattering Tomography (MST) is an effective technique for identifying special nuclear materials (SNM). Images of SNM and other materials in cargo can be produced on the basis of scattering suffered by cosmic mons while passing through the objects. In the present work, few image processing protocols are proposed with the help of simulated images for a prototype MST setup which is currently under construction. The images were produced from analyzing scattering angles within the target material using the Point of Closest Approach (PoCA) algorithm and further processed with a Pattern Recognition Method (PRM). Following this, Deep Convolutional Neural Network (DCNN) and Semantic Image Segmentation (SIS) models were employed to classify different materials.