Iqbal, Shahzeb and Zafar, Noureen and Rehman, Saif Ur and Zafar, Shireen and Mahmood, Khalid and Dzul López, Luis Alonso and García Villena, Eduardo and Ashraf, Imran UNSPECIFIED, UNSPECIFIED, UNSPECIFIED, UNSPECIFIED, UNSPECIFIED, luis.dzul@uneatlantico.es, eduardo.garcia@uneatlantico.es, UNSPECIFIED (2026) Enhancing smart city transportation networks: A novel particle swarm optimization and YOLO7x-based ensemble model for accurate vehicle detection. Ain Shams Engineering Journal, 17 (4). p. 104046. ISSN 20904479
|
Text
1-s2.0-S2090447926000730-main.pdf Download (6MB) | Preview |
Abstract
This research incorporates an intelligent transportation system (ITS) of smart cities with a newly developed ensemble model, termed EPYolov7x, through the association of particle swarm optimization (PSO) and YOLOv7x for high-accuracy traffic congestion prediction. In the proposed model, PSO is used for hyperparameter optimization for the YOLO model. Experimental results suggest a mean average precision (mAP) of 98.6% with an inference speed of 106 FPS on the UA-DETRAC dataset. It also provides good precision and real-time performance; both are essential for the practical deployment of ITS. The study findings confirm that PSO-based hyperparameter optimization improves the detection accuracy for complex traffic scenarios. The obtained results indicate superior performance compared to existing approaches. The proposed model can facilitate sustainable urban mobility toward a safer future with optimal computation.
| Item Type: | Article |
|---|---|
| Uncontrolled Keywords: | Intelligent transportation; Traffic detection; Deep learning; Particle swarm optimization; YOLOV7 |
| Subjects: | Subjects > Engineering |
| Divisions: | Europe University of Atlantic > Research > Articles and books Ibero-american International University > Research > Articles and books Ibero-american International University > Research > Articles and Books Universidad Internacional do Cuanza > Research > Articles and books University of La Romana > Research > Scientific Production |
| Depositing User: | Sr Bibliotecario |
| Date Deposited: | 17 Jul 2026 11:39 |
| Last Modified: | 17 Jul 2026 11:39 |
| URI: | https://repositorio.funiber.org/id/eprint/28812 |
Actions (login required)
 |
View Item |
