Scholarlink

The exponential vehicle growth on Indian highways has intensified the need for effective traffic monitoring. According to the 2021 National Crime Records Bureau (NCRB), India has the highest annual death toll from traffic accidents, with more than 1,55,622 people each year. Infrastructure improvements alone are insufficient, highlighting the need for a robust Intelligent Transportation System (ITS). This study aims to develop an advanced system for detecting, classifying, and tracking moving vehicles in three phases. Addressing the limitations of earlier models, such as sensitivity to lighting, noise, and traditional background subtraction, a novel method is proposed using intensity-based and pixel-orientation-aware background with fusion and multi-directional filtering. CNN models classify vehicles and detect nearby obstacles, including emergency vehicles. Object tracking is employed for speed estimation, while blindspot scanning and alerts improve driver awareness and road safety.

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

Blind Spot Aware and Speed-Sensitive Emergency Vehicle Detection Using YOLO and Faster R-CNN Synergy

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Shifa Mohammad,
  • Ashish Tiwari,
  • M. M. Dhabu

摘要

The exponential vehicle growth on Indian highways has intensified the need for effective traffic monitoring. According to the 2021 National Crime Records Bureau (NCRB), India has the highest annual death toll from traffic accidents, with more than 1,55,622 people each year. Infrastructure improvements alone are insufficient, highlighting the need for a robust Intelligent Transportation System (ITS). This study aims to develop an advanced system for detecting, classifying, and tracking moving vehicles in three phases. Addressing the limitations of earlier models, such as sensitivity to lighting, noise, and traditional background subtraction, a novel method is proposed using intensity-based and pixel-orientation-aware background with fusion and multi-directional filtering. CNN models classify vehicles and detect nearby obstacles, including emergency vehicles. Object tracking is employed for speed estimation, while blindspot scanning and alerts improve driver awareness and road safety.