Scholarlink

Drowsy driving continues to be a major cause of road accidents, and many existing monitoring systems are either too expensive or impractical for real use. In this work, we present a simple and affordable solution that runs in real time on a Raspberry Pi using MediaPipe’s facial landmark detection. By tracking key points in the eyes, mouth, and head, the system detects signs of fatigue such as prolonged eye closure, frequent yawning, and head tilt. Then it triggers alerts to keep the driver on the lookout. We tested the system in controlled settings and on a public dataset, achieving up to 100% precision on the Raspberry Pi and 70–77% precision on real-world images. The results show that even with limited hardware, reliable driver monitoring is possible.

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 Vision-Based Driver Health Monitoring System Using Raspberry Pi and Facial Landmark Analysis

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Paul Bobby Mundackal,
  • P. Muralidhar

摘要

Drowsy driving continues to be a major cause of road accidents, and many existing monitoring systems are either too expensive or impractical for real use. In this work, we present a simple and affordable solution that runs in real time on a Raspberry Pi using MediaPipe’s facial landmark detection. By tracking key points in the eyes, mouth, and head, the system detects signs of fatigue such as prolonged eye closure, frequent yawning, and head tilt. Then it triggers alerts to keep the driver on the lookout. We tested the system in controlled settings and on a public dataset, achieving up to 100% precision on the Raspberry Pi and 70–77% precision on real-world images. The results show that even with limited hardware, reliable driver monitoring is possible.