A Comparative Evaluation of YOLOv9 and DETR Models in Traffic Object Detection for Intelligent Surveillance Systems

Authors

  • Mohamad Jamil Universitas Khairun
  • Nani Nagu Universitas Khairun
  • Muhammad Said Universitas Khairun

DOI:

https://doi.org/10.29207/resti.v9i5.6556

Keywords:

intelligent systems, DETR, object detection, YOLOv9, traffic surveillance

Abstract

Object detection plays a crucial role in traffic surveillance, particularly in urban environments characterized by high vehicle density, diverse weather conditions, and limited computational resources. Although YOLOv9 and DETR have demonstrated strong performance in general object detection tasks, there is a lack of comparative research evaluating their effectiveness under specific challenges of traffic surveillance. These challenges include the need for real-time processing, accurate detection of small or partially occluded objects, and adaptability to complex traffic scenarios. This study addresses this gap by conducting a comparative evaluation of YOLOv9 and DETR using a custom traffic image dataset, with training iterations varied from 10 to 50 epochs to observe performance development. Evaluation metrics included mean average precision, precision, recall, F1-score, inference time, and object count per image. The results indicated that DETR achieved the highest accuracy across all metrics at the final training stage and detected up to 22 objects per image. However, the average inference time exceeded seven seconds per image, limiting the real-time applicability. Conversely, YOLOv9 achieved competitive accuracy with a significantly faster inference time of approximately 0.43 seconds per image. These findings provide practical insights into the trade-off between accuracy and processing efficiency, and offer guidance for model selection in operational traffic surveillance systems.

Downloads

Download data is not yet available.

References

C. Gheorghe, M. Duguleana, R. G. Boboc, and C. C. Postelnicu, ‘Analyzing Real-Time Object Detection with YOLO Algorithm in Automotive Applications: A Review’, C. - Comput. Model. Eng. Sci., vol. 141, no. 3, pp. 1939–1981, 2024, doi: 10.32604/cmes.2024.054735.

J. Le Roux and D. Van Der Haar, ‘Analysis of object detection methods to detect traffic flow’, icABCD 2019 - 2nd Int. Conf. Adv. Big Data, Comput. Data Commun. Syst., pp. 1–7, 2019, doi: 10.1109/ICABCD.2019.8851054.

R. Archana and P. S. E. Jeevaraj, Deep learning models for digital image processing: a review, vol. 57, no. 1. Springer Netherlands, 2024. doi: 10.1007/s10462-023-10631-z.

M. Ahmed, N. El-Sheimy, H. Leung, and A. Moussa, ‘Enhancing Object Detection in Remote Sensing: A Hybrid YOLOv7 and Transformer Approach with Automatic Model Selection’, Remote Sens., vol. 16, no. 1, 2024, doi: 10.3390/rs16010051.

A. O. Saltık, A. Allmendinger, and A. Stein, ‘Comparative Analysis of YOLOv9, YOLOv10 and RT-DETR for Real-Time Weed Detection’, 2024, [Online]. Available: http://arxiv.org/abs/2412.13490

Z. Jrondi, A. Moussaid, and M. Y. Hadi, ‘Exploring End-to-End object detection with transformers versus YOLOv8 for enhanced citrus fruit detection within trees’, Syst. Soft Comput., vol. 6, no. March, p. 200103, 2024, doi: 10.1016/j.sasc.2024.200103.

Y. Li, N. Miao, L. Ma, F. Shuang, and X. Huang, ‘Transformer for object detection: Review and benchmark’, Eng. Appl. Artif. Intell., vol. 126, p. 107021, 2023, doi: https://doi.org/10.1016/j.engappai.2023.107021.

A. Ćorović, V. Ilić, S. Ðurić, M. Marijan, and B. Pavković, ‘The Real-Time Detection of Traffic Participants Using YOLO Algorithm’, in 2018 26th Telecommunications Forum (TELFOR), 2018, pp. 1–4. doi: 10.1109/TELFOR.2018.8611986.

N. N. Hasibuan, M. Zarlis, and S. Efendi, ‘Detection and tracking different type of cars with YOLO model combination and deep sort algorithm based on computer vision of traffic controlling’, Sink. J. dan Penelit. Tek. Inform., vol. 6, no. 1, pp. 210–221, 2021, [Online]. Available: https://jurnal.polgan.ac.id/index.php/sinkron/article/view/11231

J. Shen, Z. Zhang, J. Luo, and X. Zhang, ‘YOLOv5-TS: Detecting traffic signs in real-time’, Front. Phys., vol. 11, no. November, pp. 1–11, 2023, doi: 10.3389/fphy.2023.1297828.

Z. Huang, X. Tao, and X. Liu, ‘NAN-DETR: noising multi-anchor makes DETR better for object detection’, Front. Neurorobot., vol. 18, 2024, doi: 10.3389/fnbot.2024.1484088.

R. Khoiriyah and A. Hendrawan, ‘Vehicle Detection on The Traffic Using Detection Transformer ( DETR ) Algorithm’, Int. J. Artif. Intell. Sci. (IJAIS, vol. 1, no. 1, pp. 14–24, 2024, doi: 10.63158/IJAIS.v1.i1.4.

Y. Wan, H. Wang, L. Lu, X. Lan, F. Xu, and S. Li, ‘An Improved Real-Time Detection Transformer Model for the Intelligent Survey of Traffic Safety Facilities’, Sustain., vol. 16, no. 23, 2024, doi: 10.3390/su162310172.

D. Nimma et al., ‘Object detection in real-time video surveillance using attention based transformer-YOLOv8 model’, Alexandria Eng. J., vol. 118, no. November 2024, pp. 482–495, 2025, doi: 10.1016/j.aej.2025.01.032.

M. Barulina, A. Andreev, I. Kovalenko, I. Barmin, E. Titov, and D. Kirillov, ‘Method for Preprocessing Video Data for Training Deep-Learning Models for Identifying Behavioral Events in Bio-Objects’, Mathematics, vol. 12, no. 24, 2024, doi: 10.3390/math12243978.

K. Maharana, S. Mondal, and B. Nemade, ‘A review : Data pre-processing and data augmentation techniques’, vol. 3, no. April, pp. 91–99, 2022, doi: 10.1016/j.gltp.2022.04.020.

A. Hassan and M. Sabha, ‘Feature Extraction for Image Analysis and Detection using Machine Learning Techniques’, Int. J. Adv. Netw. Appl., vol. 14, no. 04, pp. 5499–5508, 2023, doi: 10.35444/ijana.2023.14401.

P. V Kumaraguru and V. J. Chakravarthy, ‘An Image Feature Extraction and Image Representation Strategies for the Analysis of Image Processing’, Indian J. Forensic Med. Toxicol., vol. 11, no. 2, p. 642, 2017, doi: 10.5958/0973-9130.2017.00202.x.

E. Suherman, B. Rahman, D. Hindarto, and H. Santoso, ‘Implementation of ResNet-50 on End-to-End Object Detection (DETR) on Objects’, SinkrOn, vol. 8, no. 2, pp. 1085–1096, 2023, doi: 10.33395/sinkron.v8i2.12378.

B. Roh, J. W. Shin, W. Shin, and S. Kim, ‘Sparse Detr: Efficient End-To-End Object Detection With Learnable Sparsity’, ICLR 2022 - 10th Int. Conf. Learn. Represent., pp. 1–23, 2022.

J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, ‘You only look once: Unified, real-time object detection’, Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2016-Decem, pp. 779–788, 2016, doi: 10.1109/CVPR.2016.91.

X. Zhu, W. Su, L. Lu, B. Li, X. Wang, and J. Dai, ‘Deformable Detr: Deformable Transformers for End-To-End Object Detection’, ICLR 2021 - 9th Int. Conf. Learn. Represent., pp. 1–16, 2021.

H.-Y. S. Hao Zhang, Feng Li, Shilong Liu, Lei Zhang, Hang Su, Jun Zhu, Lionel M. Ni, ‘Dino: Detr With Improved Denoising Anchor Boxes for End-To-End Object Detection’, 11th Int. Conf. Learn. Represent. ICLR 2023, 2023, doi: 10.48550/arXiv.2203.03605.

Downloads

Published

2025-10-29

Issue

Section

Computer Science Applications

How to Cite

Jamil, M., Nagu, N., & Said, M. (2025). A Comparative Evaluation of YOLOv9 and DETR Models in Traffic Object Detection for Intelligent Surveillance Systems . Jurnal RESTI (Rekayasa Sistem Dan Teknologi Informasi), 9(5), 1283-1290. https://doi.org/10.29207/resti.v9i5.6556