International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 04 | Apr 2025
p-ISSN: 2395-0072
www.irjet.net
Railway Track Broken Detection for Train Accident Prevention Gaurav Sahu¹, Yash Yadav², Subham Kumar³, Harsh Dewangan⁴, Jay Kishan Chouhan5 Vishnu Kant Soni6 12345B.Tech Student , Department of Computer Science and Engineering , LCIT Bilaspur , CG , India
5Head of Department, Department of Computer Science and Engineering , LCIT Bilaspur , CG , India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Railway transportation continues to serve as a
This paper reviews and integrates modern techniques for railway track defect detection, including vision-based AI models, sensor systems, and geospatial monitoring tools. Emphasis is placed on technologies such as YOLOv5, IoT networks, acoustic sensors, Ground Penetrating Radar (GPR), and FPGA-based edge systems. The goal is to offer a consolidated view of how multi-modal technology can drastically enhance the effectiveness of railway safety systems.
critical backbone of logistics and public mobility worldwide. However, undetected faults in railway tracks—especially cracks and fractures—pose a major risk to operational safety. Traditional manual inspection techniques are not only labor-intensive but also prone to oversight, necessitating the adoption of smarter, automated systems. This paper presents a comprehensive study of contemporary approaches for railway track defect detection, integrating cutting-edge technologies such as sensor arrays, artificial intelligence, Internet of Things (IoT), and Ground Penetrating Radar (GPR). A detailed review of AI-based solutions like YOLOv5 for image-based crack detection, acoustic sensors for real-time fracture monitoring, and FPGA-powered edge AI platforms is provided. The paper also discusses how the fusion of geospatial mapping and GPR enhances the accuracy of subsurface defect identification. The synergy of these technologies marks a transformative step toward predictive maintenance and accident prevention, thereby ensuring safer railway operations.)
2. LITERATURE REVIEW Recent literature presents a wide array of technological interventions aimed at improving track monitoring. The following studies and innovations reflect some of the most prominent approaches: A. YOLOv5 and Geospatial Localization for Crack Detection (April 2024) A study published in April 2024 explores the combination of deep learning algorithms with spatial data for enhanced railway inspection. The system employs YOLOv5—a stateof-the-art object detection framework—for identifying cracks and defects in real-time using video and image data. What sets this method apart is its integration with GPS coordinates, allowing defects to be accurately geotagged and mapped. This significantly improves maintenance planning and reduces the time to locate and repair faults.
Key Words: Railway track monitoring, AI, YOLOv5, IoT, GPR, defect detection, accident prevention.
1. INTRODUCTION Railways are one of the most economical and widely adopted means of transportation, particularly in countries with vast geographies like India. They play a vital role in passenger travel and freight movement. However, the safety of rail operations heavily depends on the structural integrity of railway tracks. Any undetected fault—be it a surface crack, misalignment, or subsurface issue—can escalate into catastrophic accidents, leading to loss of life and substantial financial damage.
B. Acoustic Emission Sensors for Crack Detection (2017) Earlier research from 2017 focused on the application of acoustic emission (AE) technology for railway crack detection. AE sensors are capable of detecting the highfrequency waves produced by growing cracks or sudden breaks. This technique allows for continuous, real-time monitoring of rail conditions and can identify issues even in low-visibility or hard-to-reach environments.
Conventional track inspection is generally conducted through manual patrols or scheduled mechanical inspections using track recording cars. While effective to some extent, these methods are susceptible to human error, delays in defect detection, and limited frequency of checks. With the evolution of automation, artificial intelligence (AI), and real-time data communication technologies, railway systems are now transitioning toward proactive and predictive maintenance frameworks.
© 2025, IRJET
|
Impact Factor value: 8.315
C. IoT-Based Advanced Track Monitoring (May 2023) The adoption of Internet of Things (IoT) technology for railway monitoring was detailed in a 2023 study. The proposed system involved deploying smart sensors across railway infrastructure that communicated wirelessly with central databases via GSM or Wi-Fi modules. These sensors continuously monitor parameters such as vibration,
|
ISO 9001:2008 Certified Journal
|
Page 1335