NICODEME, Claire (EDF), You, Borachhun (SNCF), LY, Sovanleng (SNCF)

A Complete Framework for Automatic Condition Evaluation of Trains’ External Surface

Scheduled for presentation during the Video Session "On-Demand Video Presentations" (VP-VP), Saturday, November 22, 2025, 08:00−18:00, On-Demand Platform

2025 IEEE 28th International Conference on Intelligent Transportation Systems (ITSC), November 18-21, 2025, Gold Coast, Australia

This information is tentative and subject to change. Compiled on April 2, 2026

Abstract

Railway is a key pillar of both sustainability and transport decarbonization. Just as other industries, the railway sector is undergoing digital transformation. These new tools and innovative solutions are applied to reduce the ecological impact of rail transport even more and improve both its efficiency and effectiveness. Maintaining such complex systems requires a lot of resources, and intensive and cautious manual work. Computer Vision and AI may bring additional insight to improve and make the maintenance processes more robust. This paper presents a video analytics framework, to detect and identify a train, while applying condition monitoring algorithms to various elements. First, the method proposes to generate the train’s panoramic image from event-based acquisition, giving a global view of the vehicle. Then logos, texts and liveries are analyzed to ensure regulatory conformity, at both European and national levels. The designed pipeline aims at faster and more reliable maintenance, while also improving defects detection. Results on real-world data demonstrate the effectiveness of visual elements recognition and defects detection.