Scholarlink

In traditional classrooms, attendance tracking is often time-consuming and error-prone, relying on manual methods like roll call and sign-in sheets. This paper presents an Automated Attendance System designed specifically for students at Vietnam National University International School (VNU-IS), utilizing face recognition technology to streamline the process. Our system employs Multi-task Cascaded Convolutional Networks (MTCNN) for face detection and alignment and FaceNet for feature extraction and recognition. The recognized faces are classified using Support Vector Machines (SVM), and attendance records are managed with SQLite and Excel for real-time updates. The system was evaluated under various lighting conditions and facial angles, demonstrating high precision and recall, particularly in well-lit environments. This approach enhances the accuracy and efficiency of attendance tracking, reduces manual intervention, and provides valuable insights for educators and administrators.

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

Building an Automatic Attendance System for VNUIS Students Using Convolutional Neural Network

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Thai Dinh Kim,
  • D. T. Phuong-Thao,
  • Manh-Hung Ha,
  • Anh-Phuong Pham,
  • Khac-Ton Nguyen,
  • Ngoc-Nam Dao,
  • Khac-Truong Nguyen,
  • Ha Nguyen-Tien

摘要

In traditional classrooms, attendance tracking is often time-consuming and error-prone, relying on manual methods like roll call and sign-in sheets. This paper presents an Automated Attendance System designed specifically for students at Vietnam National University International School (VNU-IS), utilizing face recognition technology to streamline the process. Our system employs Multi-task Cascaded Convolutional Networks (MTCNN) for face detection and alignment and FaceNet for feature extraction and recognition. The recognized faces are classified using Support Vector Machines (SVM), and attendance records are managed with SQLite and Excel for real-time updates. The system was evaluated under various lighting conditions and facial angles, demonstrating high precision and recall, particularly in well-lit environments. This approach enhances the accuracy and efficiency of attendance tracking, reduces manual intervention, and provides valuable insights for educators and administrators.