Scholarlink

The Drowsiness Detection model presented in this research work utilizes an advanced Convolutional Neural Network (CNN) model to monitor driver alertness through real-time eye state analysis. By utilizing deep learning and computer vision, it detects early signs of drowsiness by tracking eye closure duration and frequency, providing a non-intrusive, highly accurate solution to mitigate fatigue-related accidents. The model architecture is based on InceptionV3, optimized with custom layers to enhance the classification accuracy for eye states (open or closed). Real-time functionality is supported through OpenCV and Pygame for continuous video monitoring and immediate audio alerts.

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

A Comparative Study of CNN Models for Real-Time Driver Drowsiness Detection

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Ayush Shah,
  • Harsh Giri,
  • Bhavya Nirwan,
  • Shola Usharani

摘要

The Drowsiness Detection model presented in this research work utilizes an advanced Convolutional Neural Network (CNN) model to monitor driver alertness through real-time eye state analysis. By utilizing deep learning and computer vision, it detects early signs of drowsiness by tracking eye closure duration and frequency, providing a non-intrusive, highly accurate solution to mitigate fatigue-related accidents. The model architecture is based on InceptionV3, optimized with custom layers to enhance the classification accuracy for eye states (open or closed). Real-time functionality is supported through OpenCV and Pygame for continuous video monitoring and immediate audio alerts.