Real-Time Railway Track Defect Detection Using an Optimized YOLOv8 Framework

Abstract

Railway track defect detection is vital for ensuring 
the safety and reliability of train operations and 
infrastructure. Traditional defect detection methods 
often suffer from missed detections, inaccurate 
localization, and limited capability in identifying 
small-scale anomalies, particularly under complex 
environmental conditions. To overcome these 
limitations, this work proposes an improved track 
defect detection network based on YOLOv8, referred