A Comprehensive Review of Deep Learning and Computer Vision Approaches for Railway Track Crack Detection

Harivansh Markam; Nikita Rawat

A Comprehensive Review of Deep Learning and Computer Vision Approaches for Railway Track Crack Detection

Author(s)	Mr. Harivansh Markam, Ms. Nikita Rawat
Country	India
Abstract	Railway track integrity is a fundamental determinant of rail safety worldwide. The timely detection of surface cracks, structural fractures, and rail head defects can prevent catastrophic derailments and save lives. Over the past six years, research in automated track inspection has undergone a paradigm shift — moving from threshold-based sensor systems and handcrafted image processing pipelines to sophisticated deep learning architectures achieving near-human detection accuracy. This review paper surveys and critically analyses twenty significant research contributions on railway track crack detection and monitoring, spanning 2019–2025. The reviewed works are categorized into four modalities: (i) sensor-based IoT systems, (ii) classical image processing approaches, (iii) deep learning and convolutional neural network methods including object detection architectures such as YOLOv5 and transformer-based models, and (iv) hybrid multi-modal systems. Key open challenges including dataset standardisation, edge deployment of deep learning models, multi-class severity grading, and integration with digital twin infrastructure management platforms are identified, and a research roadmap for the 2025–2030 period is proposed.
Keywords	Railway track crack detection, deep learning, YOLOv5, MobileNetV2, transfer learning, computer vision, object detection, convolutional neural network, Streamlit, IoT, non-destructive testing, review
Field	Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In	Volume 8, Issue 3, May-June 2026
Published On	2026-06-12

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

A Comprehensive Review of Deep Learning and Computer Vision Approaches for Railway Track Crack Detection

Share this