DEVELOPING A SYSTEM FORCRACK CLASSIFICATION IN TRAIN TRACKSUSING MACHINE LEARNING

The proposed system “Developing a System for Crack Classification in Train Tracks using Machine Learning” focuses on improving railway safety by automatically detecting cracks in railway tracks. Railway tracks are continuously exposed to heavy loads, weather conditions, and constant train movement, which may cause cracks or structural damage over time. If these cracks are not detected at an early stage, they may lead to serious railway accidents and safety issues.   In this project, machine learning and image processing techniques are used to analyze railway track images and detect cracks effectively. The system collects railway track images, preprocesses them to improve image quality, and extracts important features such as crack patterns and edges.  These features are then used to train a machine learning model such as a Convolutional Neural Network (CNN).

1.  J.-Y. Choi and J.-M. Han, “Deep Learning (Fast R-CNN)-Based Evaluation of Rail Surface Defects,” Applied Sciences, 2024.   2. S. Sowmya, R. Surendhar, K. Karthikeyan, “A Novel Machine Learning Framework for Real-Time Railway Crack and Squat Detection,” IJRASET, 2025.   3.S. Pradeep Kumar et al., “Artificial Intelligence-Based Smart Automatic Railway Crack Detection and Protection,” 2026.   4. J. Zhang et al., “AI-Optimized Ultrasonic Guided Wave Modes for Railway Crack Detection,” Biomimetic Intelligence and Robotics, 2024.The paper focuses on using AI algorithms to optimize ultrasonic guided waves for accurate crack detection and localization in railway tracks. 5.  J.-Y. Choi, “Rail Surface Defect Detection Using Deep Learning,” 2024. This study proposes deep learning methods for identifying rail surface defects and improving inspection efficiency.   6.  S. Rathi et al., “Deep Learning Solutions for Automated Crack Detection in Metallic Surfaces,” International Journal of Intelligent Systems and Applications, 2024. 7.  M. Shaikh et al., “FaultSeg Dataset for Train Wheel Defect Detection,” Scientific Data, 2025. The study introduces a dataset designed for defect detection in railway components using machine learning.   8.  Y. Cha, W. Choi, and O. Büyüköztürk, “Deep Learning-Based Crack Damage Detection Using Convolutional Neural Networks,” Computer-Aided Civil and Infrastructure Engineering, 2017. 9.  H. Li, D. Song, and Y. Liu, “Automatic Pixel-Level Crack Detection Using Deep Convolutional Neural Networks,” IEEE Access, 2019. 10.  Q. Zou, Y. Cao, Q. Li, Q. Mao, and S. Wang, “CrackTree: Automatic Crack Detection from Pavement Images,” Pattern Recognition Letters, 2012.