Railway Crack Detection Robot Dataset

Overview

• Source: Manually collected
• Content: Images of railway tracks both with and without cracks
• Processing: Proper annotation and pre‑processing steps were applied to guarantee dataset quality

Model Training

• Model Type: Convolutional Neural Network (CNN)
• Training Data: The manually collected dataset
• Training Script: Use the model.py script for training or testing

Real‑Time Detection

• Realtime Script: Use the realtime.py script for live crack detection
• Configuration: Ensure the camera or video input is correctly configured