Railway Track Crack Detection System Using Deep Learning and Computer Vision

Harivansh Markam; Nikita Rawat

Railway Track Crack Detection System Using Deep Learning and Computer Vision

Author(s)	Mr. Harivansh Markam, Ms. Nikita Rawat
Country	India
Abstract	Railway infrastructure is a critical component of national transportation systems, and maintaining track integrity is paramount for ensuring passenger and cargo safety. Track defects — including surface cracks, transverse fractures, and structural deformations — are a leading cause of derailments worldwide. This paper presents a Railway Track Crack Detection System employing a dual-model deep learning approach: a MobileNetV2-based transfer learning classifier for binary defect classification, and a YOLOv5 object detection model deployed via the Roboflow Inference API for spatial localization with annotated bounding boxes. The MobileNetV2 model uses a two-phase training strategy with seven data augmentation techniques to improve generalization. The complete system is deployed as an interactive web application using Streamlit, enabling end-to-end inference where a user uploads a track image and receives annotated detection results with confidence scores in real time. Experimental results demonstrate high accuracy, precision, recall, and F1-score, significantly outperforming classical feature engineering baselines.
Keywords	Railway track crack detection, deep learning, YOLOv5, MobileNetV2, transfer learning, computer vision, Streamlit, object detection, convolutional neural network, predictive maintenance
Field	Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In	Volume 8, Issue 3, May-June 2026
Published On	2026-06-04

View / Download PDF File

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Railway Track Crack Detection System Using Deep Learning and Computer Vision

Share this