Scholarlink

<p>Driver sleepiness is one of the causes of automobile accidents. This is caused by a lack of sleep, fatigue, lethargy, exhaustion, tiredness, and other physical problems. This is a prevalent issue among long-distance drivers, which leads to accidents. Avoiding drowsy driving accidents requires forewarning or alerting the driver ahead of time. Several methods have been used for diagnosing tiredness. The face, being a vital aspect of the body, provides a large amount of information. In a proposed work, various machine learning techniques are applied to identify driver fatigue, including support vector machine, random forest, and k-nearest neighbour for detecting driver drowsiness (DD). The Histogram of Oriented Gradient (HOG) algorithm is employed to compare the normal and drowsy facial images. As a vital region of the human body, the face conveys a wealth of information relevant to such analysis. When a driver becomes fatigued, facial expressions—such as increased blinking rate, and frequent yawning, and prolonged eye closure – deviate from their normal patterns. The proposed method integrates real-time video frames and reference images to evaluate the DD based on these behavioral indicators. The proposed “Drowsy Driving Warning System” aims to prevent accidents and enhance road safety. It utilizes open- and closed-eye images to predict the driver’s level of drowsiness.</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 algorithm-based driver drowsiness detection system

  • 期刊论文

  • 发表日期：

    2025年7月30日

查看全文
  • Sumathi M,
  • Peddi Sathwik,
  • Karakala Bhanu Prakash Reddy,
  • Sai Sreekar M,
  • S. P. Raja

摘要

Driver sleepiness is one of the causes of automobile accidents. This is caused by a lack of sleep, fatigue, lethargy, exhaustion, tiredness, and other physical problems. This is a prevalent issue among long-distance drivers, which leads to accidents. Avoiding drowsy driving accidents requires forewarning or alerting the driver ahead of time. Several methods have been used for diagnosing tiredness. The face, being a vital aspect of the body, provides a large amount of information. In a proposed work, various machine learning techniques are applied to identify driver fatigue, including support vector machine, random forest, and k-nearest neighbour for detecting driver drowsiness (DD). The Histogram of Oriented Gradient (HOG) algorithm is employed to compare the normal and drowsy facial images. As a vital region of the human body, the face conveys a wealth of information relevant to such analysis. When a driver becomes fatigued, facial expressions—such as increased blinking rate, and frequent yawning, and prolonged eye closure – deviate from their normal patterns. The proposed method integrates real-time video frames and reference images to evaluate the DD based on these behavioral indicators. The proposed “Drowsy Driving Warning System” aims to prevent accidents and enhance road safety. It utilizes open- and closed-eye images to predict the driver’s level of drowsiness.