Jorge Luis Díaz-Acosta
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Jorge Luis Díaz-Acosta has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Cándido Caballero-Gil
Department of Computer Science and Systems, University of La Laguna, 38270 La Laguna, Spain
1 shared publication - Pino Caballero-Gil
Universidad de La Laguna
1 shared publication Rubén Franco-Villa
Metropolitano de Tenerife S.A., San Cristbal de La Laguna, Spain
1 shared publication- Sonia Díaz-Santos
Department of Computer Engineering and Systems, University of La Laguna, San Cristbal de La Laguna, Spain
1 shared publication
Latest Publications
- Open Access
- 6 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. PP (15249050) , 1-23
Deep Learning-Based Obstacle Detection and Traffic Light Recognition in Tram Systems Using Jetson Nano
This research explores the development and implementation of an artificial intelligence (AI)-based system for obstacle detection and traffic light recognition in the tram network of Tenerife, Canary Islands, Spain. The motivation for this study lies in the need to improve safety and operational efficiency in public transport systems through the use of current object detection technologies. For this purpose, a comprehensive review of current AI-based object detection technologies is carried out, with a focus on deep learning methods. As a result of this analysis, a You Only Look Once (YOLO) model, optimized using transfer learning, is selected and implemented in a real-time software application running on a Jetson Nano device. This system allows fast and accurate detection of obstacles and traffic signals, providing immediate alerts to operators. The objective of the system is to design and evaluate a system capable of assisting trams operators in identifying potential risks in real time through timely alerts, contributing to accident reduction and service optimization. The results obtained demonstrate the feasibility and effectiveness of using deep learning techniques to strengthen safety and operational efficiency in trams. Furthermore, this study demonstrates how advanced computer vision techniques can be successfully applied in public transportation environments, laying the groundwork for future improvements in driver assistance systems. This research contributes to the field of AI in transportation by demonstrating a practical application of these technologies in public transit systems, highlighting their potential to significantly improve driving assistance systems.
- 6 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. PP (15249050) , 1-23
Open AccessDeep Learning-Based Obstacle Detection and Traffic Light Recognition in Tram Systems Using Jetson Nano
This research explores the development and implementation of an artificial intelligence (AI)-based system for obstacle detection and traffic light recognition in the tram network of Tenerife, Canary Islands, Spain. The motivation for this study lies in the need to improve safety and operational efficiency in public transport systems through the use of current object detection technologies. For this purpose, a comprehensive review of current AI-based object detection technologies is carried out, with a focus on deep learning methods. As a result of this analysis, a You Only Look Once (YOLO) model, optimized using transfer learning, is selected and implemented in a real-time software application running on a Jetson Nano device. This system allows fast and accurate detection of obstacles and traffic signals, providing immediate alerts to operators. The objective of the system is to design and evaluate a system capable of assisting trams operators in identifying potential risks in real time through timely alerts, contributing to accident reduction and service optimization. The results obtained demonstrate the feasibility and effectiveness of using deep learning techniques to strengthen safety and operational efficiency in trams. Furthermore, this study demonstrates how advanced computer vision techniques can be successfully applied in public transportation environments, laying the groundwork for future improvements in driver assistance systems. This research contributes to the field of AI in transportation by demonstrating a practical application of these technologies in public transit systems, highlighting their potential to significantly improve driving assistance systems.