RADS: Railtrack Anomaly Detection System Using YOLOv8 for Automated Railway Track Inspection

One of the most common ways to get around the world is by train. The structural soundness of rail tracks is very important for the safety and efficiency of railway systems. Cracks, fractures, and misalignment in tracks can cause serious accidents and problems with operations. Most traditional ways to inspect railways are done by hand and take a lot of time and people. These methods can also be wrong because of people and not enough inspections. This study presents RADS (Railtrack Anomaly Detection System), a deep learning-based framework engineered to autonomously identify defects in railway tracks through computer vision. The suggested system uses the YOLOv8 object detection architecture to find strange things in railtrack images. A dataset with over 1000 pictures of rail tracks was used to teach and test the model. These images include both defective and non-defective track conditions and were manually annotated using Roboflow to ensure high-quality labeling.