Scholarlink

Abstract <p>This study explores the incorporation of specialized self-attention mechanisms into deep learning architectures, with a particular emphasis on segmenting human iris and pupil regions in infrared images. In&#xa0;this work, we present some modified versions of nonlocal blocks designed to enhance self-attentive properties while addressing the distinct characteristics of infrared imaging data. By applying these customized enhancements, we have observed significant improvements in the performance metrics of the underlying deep neural network model, with segmentation results showing substantial improvement (increasing from 0.945 to 0.983 in mIoU and from 0.951 to 0.988 in mDice) after thorough testing on a representative infrared image dataset. This advancement paves the way for various applications in infrared image analysis, offering new opportunities for research and innovation.</p>

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

Modernized Nonlocal Blocks for Infrared Camera Image Segmentation of the Human Eye

  • 期刊论文

  • 发表日期：

    2025年6月25日

查看全文
  • Aleksei Samarin,
  • Alexander Savelev,
  • Aleksei Toropov,
  • Artem Nazarenko,
  • Alexander Motyko,
  • Elena Mikhailova,
  • Egor Kotenko,
  • Alina Dzestelova,
  • Valentin Malykh

摘要

Abstract

This study explores the incorporation of specialized self-attention mechanisms into deep learning architectures, with a particular emphasis on segmenting human iris and pupil regions in infrared images. In this work, we present some modified versions of nonlocal blocks designed to enhance self-attentive properties while addressing the distinct characteristics of infrared imaging data. By applying these customized enhancements, we have observed significant improvements in the performance metrics of the underlying deep neural network model, with segmentation results showing substantial improvement (increasing from 0.945 to 0.983 in mIoU and from 0.951 to 0.988 in mDice) after thorough testing on a representative infrared image dataset. This advancement paves the way for various applications in infrared image analysis, offering new opportunities for research and innovation.