본문 바로가기
  • Home

A Scheme Pedestrian Accident Prevention System for Right Turn using YOLOv8x

  • Journal of Internet of Things and Convergence
  • Abbr : JKIOTS
  • 2024, 10(6), pp.117-123
  • Publisher : The Korea Internet of Things Society
  • Research Area : Engineering > Computer Science > Internet Information Processing
  • Received : November 8, 2024
  • Accepted : December 10, 2024
  • Published : December 31, 2024

Seok-Jin Kim 1 park sang min 2 Chan-Hwi Lee 1 Se-Young Jang 1 Woo-Hyuk Jang 1 Su-Min Joo 1 Keun-Ho Lee 3

1백석대학교 컴퓨터공학부
2백석대학교
3백석대학교 정보통신학부

Accredited

ABSTRACT

In the past, making a right turn at intersections relied entirely on the driver's judgment. The Act on the Passage of Intersections was revised in 2023 to require drivers to yield or pause when there are pedestrians crossing or about to cross the crosswalk. Despite the revision, many drivers are unaware of the new law and continue to make right turns as before. Moreover, some drivers have a complacent attitude, thinking "nothing will happen anyway" when they follow the old method. Even drivers who know the correct way to pass often make sudden stops due to pedestrians being in blind spots. This creates dangerous situations for pedestrians. This paper proposes a system that warns drivers about pedestrians' presence using the YOLOv8x model. The system is activated based on the presence of pedestrians, creating a light induction effect that helps drivers naturally recognize pedestrians, preventing accidents.

Citation status

* References for papers published after 2023 are currently being built.