Scholarlink

<p>Emotions are vital in education, influencing learning outcomes significantly. This paper models facial expressions using Partial Differential Equations (PDEs) to capture the geometry of facial features. A fourth-order elliptical PDE approximates generic face structure via spectral solutions. An IGWO + AEISOM approach is proposed for effective emotion recognition and feature optimization. IGWO enhances AEISOM by eliminating redundant features and boosting classification accuracy. An Automated Facial Expression Recognition (AFER) system is integrated into online learning environments. Facial images are captured and categorized into six emotions using the AEISOM classifier. Evaluations on the CK + , JAFFE, and FER-2013 datasets show superior accuracy of 99.4%, 99.6%, and 94.3%, respectively.</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

Automatic facial expression recognition based on improved grey wolf optimization algorithm with AEISOM classifier

  • 期刊论文

  • 发表日期：

    2025年9月11日

查看全文
  • Raavi Deepthi,
  • Swapna Siddamsetti,
  • Deepika Mallampati,
  • Pilligundla Niharika,
  • T. Raghunadha Reddy

摘要

Emotions are vital in education, influencing learning outcomes significantly. This paper models facial expressions using Partial Differential Equations (PDEs) to capture the geometry of facial features. A fourth-order elliptical PDE approximates generic face structure via spectral solutions. An IGWO + AEISOM approach is proposed for effective emotion recognition and feature optimization. IGWO enhances AEISOM by eliminating redundant features and boosting classification accuracy. An Automated Facial Expression Recognition (AFER) system is integrated into online learning environments. Facial images are captured and categorized into six emotions using the AEISOM classifier. Evaluations on the CK + , JAFFE, and FER-2013 datasets show superior accuracy of 99.4%, 99.6%, and 94.3%, respectively.