Synthetic Data-Driven Object Detection for Rail Transport: YOLO vs RT-DETR in Train Loading Operations

Thiago Leonardo Maria, Saul Delabrida, Saul Delabrida, Andrea Gomes Campos, Andrea Gomes Campos

2025

Abstract

Efficient wagon loading plays a crucial role in logistic efficiency and supplying essential raw materials to various industries. However, ensuring the cleanliness of the wagons before loading is a critical aspect of this process as it directly impacts the quality and integrity of the transported item. Early detection of objects inside empty wagons before loading is a key component in this logistic puzzle. This study proposes a computer vision approach for object detection in train wagons before loading and performs a comparison between two models: YOLO (You Only Look Once) and RT-DETR (Real-Time Detection Transformer), which are based on Convolutional Neural Networks (CNNs) and Transformers, respectively. Additionally, the research addresses the generation of synthetic data as a strategy for model training, using the \textit{Unity} platform to create virtual environments that simulate real conditions of wagon loading. Therefore, the findings highlight the potential of combining computer vision and synthetic data to improve the safety, efficiency, and automation of train loading processes, offering valuable insights into the application of advanced vision models in industrial scenarios.

Paper Citation

in Harvard Style

Maria T., Delabrida S. and Campos A. (2025). Synthetic Data-Driven Object Detection for Rail Transport: YOLO vs RT-DETR in Train Loading Operations. In Proceedings of the 27th International Conference on Enterprise Information Systems - Volume 1: ICEIS; ISBN 978-989-758-749-8, SciTePress, pages 873-880. DOI: 10.5220/0013402500003929

in Bibtex Style

@conference{iceis25,
author={Thiago Maria and Saul Delabrida and Andrea Campos},
title={Synthetic Data-Driven Object Detection for Rail Transport: YOLO vs RT-DETR in Train Loading Operations},
booktitle={Proceedings of the 27th International Conference on Enterprise Information Systems - Volume 1: ICEIS},
year={2025},
pages={873-880},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013402500003929},
isbn={978-989-758-749-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 27th International Conference on Enterprise Information Systems - Volume 1: ICEIS
TI - Synthetic Data-Driven Object Detection for Rail Transport: YOLO vs RT-DETR in Train Loading Operations
SN - 978-989-758-749-8
AU - Maria T.
AU - Delabrida S.
AU - Campos A.
PY - 2025
SP - 873
EP - 880
DO - 10.5220/0013402500003929
PB - SciTePress