AI Driven railway crack detection system using Convolutional Neural Network and IoT
| dc.contributor.author | Sudha, M. | |
| dc.contributor.author | Saranya, R. | |
| dc.contributor.author | Ali, Guma | |
| dc.contributor.author | Ganesh, C | |
| dc.contributor.author | Umamaheswari, S. | |
| dc.date.accessioned | 2025-12-18T15:50:06Z | |
| dc.date.available | 2025-12-18T15:50:06Z | |
| dc.date.issued | 2025-06-02 | |
| dc.description | This study presents an AI- and IoT‑enabled railway crack‑detection system that automates fracture identification in railway tracks, aimed at enhancing safety and reducing manual inspection efforts. It advances SDG 9 by fostering industry innovation and resilient infrastructure, and SDG 11 by contributing to safer, more sustainable transport systems. While the work doesn’t directly target SDG 6, its infrastructure focus supports broader development frameworks. The proposed solution aligns with Uganda’s NDP IV aspirations for infrastructure enhancement and industrialization, offering a practical model for modernizing transport assets, improving public safety, and supporting economic transformation through technological adoption. | |
| dc.description.abstract | The Railway Track fractures identification Using AI with IoT project aims to increase railway safety by automating the identification of fractures in railway tracks using artificial intelligence (AI) and Internet of Things (IoT) technology. Even minor cracks in railway tracks, which are crucial components of the transportation system, can cause major accidents if they not immediately identified also repaired. Conventional manual assessing technique are expensive, time-consuming, and error-prone. Convolutional Neural Networks, a kind deep learning model, will used in this study to automatically identify cracks in photos of railroad tracks. The model trained on a dataset of track photographs in order to discern parts are defective (cracked) and non-defective (intact). Once trained, The CNN model is employed to analyze images captured by cameras mounted on trains or inspection vehicles as part of a real-time monitoring system driven by the Internet of Things. Every time a fracture is discovered, the system sends the information to the Blynk IoT platform, notifying and alerting maintenance personnel. Additionally, an LCD display and a buzzer alarm are activated in the field to alert technicians to the detected defect. By combining AI and IoT, the initiative aims to reduce overall maintenance costs, improve safety through early fault detection, and speed up the track inspection process. By providing a more automated, accurate, and efficient method of tracking the present state of railroad lines, the system ultimately increases the harmlessness and reliability of railway transit. | |
| dc.identifier.citation | Sudha, M., Ali, G., Ganesh, C., Umamaheswari, S., & Saranya, R. (2025, April). AI Driven railway crack detection system using Convolutional Neural Network and IoT. In 2025 3rd International Conference on Advancements in Electrical, Electronics, Communication, Computing and Automation (ICAECA): 1-8. | |
| dc.identifier.uri | https://dir.muni.ac.ug/handle/20.500.12260/841 | |
| dc.publisher | IEEE | |
| dc.subject | Transportation | |
| dc.subject | Railway safety | |
| dc.subject | Inspection | |
| dc.subject | Rail transportation | |
| dc.subject | Liquid crystal displays | |
| dc.subject | Real-time systems | |
| dc.subject | Maintenance | |
| dc.subject | Safety | |
| dc.subject | Convolutional neural networks | |
| dc.subject | Internet of Things | |
| dc.title | AI Driven railway crack detection system using Convolutional Neural Network and IoT | |
| dc.type | Other |