Scholarlink

<p>Fatigued drivers often cause traffic accidents. This study introduces a novel method for detecting fatigue that combines machine learning and image processing techniques. We propose a unique approach that utilizes the Haar Cascade method, the CatBoost algorithm, and an Inception V3 for facial detection and eye classification, allowing for the quick identification and management of drowsiness-related issues. The proposed method uses Haar Cascade for facial detection, proving more reliable than CNN-based methods. A Python-based program detects the real-time images of the driver’s face using OpenCV, while deep learning models built on Keras analyze the facial data. The CNN is trained to distinguish between open and closed eyes, aiding in detecting fatigue. When prolonged eye closure is detected, drivers receive immediate advice to stop or take a break. This effort aims to create a fatigue detection system that is both reliable and robust, capable of swiftly identifying prolonged eye closure. Ultimately, our method has the potential to improve road safety significantly and contribute to global initiatives aimed at addressing this critical issue by providing early warnings to drivers.</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

Machine learning-based augmented vision for detecting driver drowsiness

  • 期刊论文

  • 发表日期：

    2025年4月8日

查看全文
  • Gude Ramarao,
  • Palangthod Soni,
  • Agraharam Sri Vaishnavi,
  • VenkataVarshitha

摘要

Fatigued drivers often cause traffic accidents. This study introduces a novel method for detecting fatigue that combines machine learning and image processing techniques. We propose a unique approach that utilizes the Haar Cascade method, the CatBoost algorithm, and an Inception V3 for facial detection and eye classification, allowing for the quick identification and management of drowsiness-related issues. The proposed method uses Haar Cascade for facial detection, proving more reliable than CNN-based methods. A Python-based program detects the real-time images of the driver’s face using OpenCV, while deep learning models built on Keras analyze the facial data. The CNN is trained to distinguish between open and closed eyes, aiding in detecting fatigue. When prolonged eye closure is detected, drivers receive immediate advice to stop or take a break. This effort aims to create a fatigue detection system that is both reliable and robust, capable of swiftly identifying prolonged eye closure. Ultimately, our method has the potential to improve road safety significantly and contribute to global initiatives aimed at addressing this critical issue by providing early warnings to drivers.