Scholarlink

Potholes detection is crucial for road infrastructure. These potholes are considers as significant threat to any street structure, leading to increased maintenance costs, traffic disruptions, and safety hazards. This research paper explores the application of the You Only Look Once (YOLO) algorithms for pothole detection, a computer vision technique renowned for its real time object detection capabilities. The study aims to evaluate and compare various versions of the YOLO algorithm to ascertain their effectiveness in accurately detecting potholes from images and video streams. The selected YOLO versions for comparison include YOLOv5, YOLOv7 and YOLOv8 Through a comprehensive analysis, this research seeks to provide insights into the strengths and weaknesses of each version, shedding light on their potential applications for pothole detection in road maintenance and management.

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

Optimizing Pothole Detection Using YOLO Variants: v5, v7, and v8 Performance Benchmark

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Sana Fatima,
  • Tanazzah Rehman,
  • Najmi Ghani Haider,
  • Moaaz Siddiqui

摘要

Potholes detection is crucial for road infrastructure. These potholes are considers as significant threat to any street structure, leading to increased maintenance costs, traffic disruptions, and safety hazards. This research paper explores the application of the You Only Look Once (YOLO) algorithms for pothole detection, a computer vision technique renowned for its real time object detection capabilities. The study aims to evaluate and compare various versions of the YOLO algorithm to ascertain their effectiveness in accurately detecting potholes from images and video streams. The selected YOLO versions for comparison include YOLOv5, YOLOv7 and YOLOv8 Through a comprehensive analysis, this research seeks to provide insights into the strengths and weaknesses of each version, shedding light on their potential applications for pothole detection in road maintenance and management.