Performance Comparison of YOLOv8n and Faster R-CNN for Automatic Detection of School Uniform Attributes

  • Argi Ginanjar Universitas Nusa Putra ORCID
  • Adhi Kusnadi Program Magister Informatika, Universitas Nusa Putra, Sukabumi, Indonesia
  • Risnandar Program Magister Informatika, Universitas Nusa Putra, Sukabumi, Indonesia
PDF
DOI https://doi.org/10.31544/jtera.v11.i1.2026.189-198
Published 2026-06-26
Full Text Preview
Download PDF

Abstract

Manual inspection of students’ uniform completeness is still commonly performed in schools, especially in vocational schools with a large number of students. This process is time-consuming, prone to subjectivity, and difficult to document consistently. This study aims to compare the performance of YOLOv8n and Faster R-CNN MobileNetV3 320 FPN in automatically detecting school uniform attributes. The dataset used in this study consists of 638 school uniform images with five object classes, namely epolet, name_tag, sabuk, sepatu_hitam, and topi. Both models were trained using the same dataset and evaluated based on Intersection over Union (IoU), mean Average Precision (mAP), and Frames per Second (FPS) under the experimental configuration used in this study. The experimental results show that YOLOv8n achieved an IoU of 0.8713, mAP50 of 0.9193, mAP50-95 of 0.6695, and an inference speed of 83.5921 FPS. Meanwhile, Faster R-CNN achieved an IoU of 0.6714, mAP50 of 0.6640, mAP50-95 of 0.3853, and an inference speed of 9.5801 FPS. These results indicate that, under the experimental configuration and evaluation subsets used in this study, YOLOv8n provides a better balance between localization accuracy, detection accuracy, and inference efficiency. Therefore, YOLOv8n has the potential to be more suitable as a basis for developing a semi-real-time school uniform completeness monitoring system.

References

[1] J. Terven, D. M. Córdova-Esparza, and J. A. Romero-González, “A Comprehensive Review of YOLO Architectures in Computer Vision: From YOLOv1 to YOLOv8 and YOLO-NAS,” Mach. Learn. Knowl. Extr., vol. 5, no. 4, pp. 1680–1716, 2023, doi: 10.3390/make5040083.
[2] M. L. Ali and Z. Zhang, “The YOLO framework: A comprehensive review of evolution, applications, and benchmarks in object detection,” Computers, vol. 13, no. 12, p. 336, 2024.
[3] Z. Zou, K. Chen, Z. Shi, Y. Guo, and J. Ye, “Object detection in 20 years: A survey,” Proc. IEEE, vol. 111, no. 3, pp. 257–276, 2023.
[4] M. Trigka and E. Dritsas, “A Comprehensive Survey of Machine Learning Techniques and Models for Object Detection,” Sensors, vol. 25, no. 1, 2025, doi: 10.3390/s25010214.
[5] Q. Feng, X. Xu, and Z. Wang, “Deep learning-based small object detection: A survey,” Math. Biosci. Eng., vol. 20, no. 4, pp. 6551–6590, 2023.
[6] A. Aldubaikhi and S. Patel, “Advancements in small-object detection (2023–2025): Approaches, datasets, benchmarks, applications, and practical guidance,” Appl. Sci., vol. 15, no. 22, p. 11882, 2025.
[7] Y.-H. Chang and Y.-Y. Zhang, “Deep learning for clothing style recognition using YOLOv5,” Micromachines, vol. 13, no. 10, p. 1678, 2022.
[8] X. Li, “Attention-Driven Multi-Scale Clothing Detection Using an Enhanced SCS-YOLO Framework,” Int. J. Intell. Inf. Technol., vol. 21, no. 1, pp. 1–19, 2025.
[9] R. Padilla, W. L. Passos, T. L. B. Dias, S. L. Netto, and E. A. B. Da Silva, “A comparative analysis of object detection metrics with a companion open-source toolkit,” Electronics, vol. 10, no. 3, p. 279, 2021.
[10] S. Paniego, V. Sharma, and J. M. Cañas, “Open source assessment of deep learning visual object detection,” Sensors, vol. 22, no. 12, p. 4575, 2022.
[11] PyTorch, “fasterrcnn_mobilenet_v3_large_320_fpn,” May 19, 2026, PyTorch. [Online]. Available: https://docs.pytorch.org/vision/master/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html
[12] N. A. Megantara and E. Utami, “Object Detection using YOLOv8: A Systematic Review,” Sist. J. Sist. Inf., vol. 14, no. 3, pp. 1186–1193, 2025.
[13] H. Zhang and X. Yang, “A Survey on Algorithms and Performance Metrics for Object Detection,” Int. Core J. Eng., vol. 9, no. 9, pp. 133–145, 2023.
[14] M. S. Amin, C. Wang, and S. Jabeen, “Fashion sub-categories and attributes prediction model using deep learning,” Vis. Comput., vol. 39, no. 9, pp. 3851–3864, 2023.
[15] V. Parekh and K. Shaik, “Multi-Scale Feature Based Fashion Attribute Extraction Using Multi-Task Learning for e-Commerce Applications.,” in VISIGRAPP (5: VISAPP), 2023, pp. 644–651.
[16] Y. Wan, C. Yan, B. Zhang, and G. Zou, “Learning image representation via attribute-aware attention networks for fashion classification,” in International Conference on Multimedia Modeling, Springer, 2022, pp. 69–81.
[17] A. F. Saputra and Y. Anang, “Development of Student’s Uniform Compliance Detection System Using Real Time Image Recognition at Politeknik Statistika STIS,” in Proceedings of The International Conference on Data Science and Official Statistics, 2023, pp. 58–72.
[18] M. Dody and Y. Yohannes, “Perangkat Lunak Pendeteksi Jenis Seragam Siswa Jenjang Pendidikan Menengah Menggunakan YOLOv8,” JATISI, vol. 12, no. 2, 2025.
[19] V. Parekh, K. Shaik, S. Biswas, and M. Chelliah, “Fine-grained visual attribute extraction from fashion wear,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2021, pp. 3973–3977.
[20] O. Nocentini, J. Kim, M. Z. Bashir, and F. Cavallo, “Image classification using multiple convolutional neural networks on the fashion-MNIST dataset,” Sensors, vol. 22, no. 23, p. 9544, 2022.

Issue

Vol. 11 No. 1: June 2026

Section

Articles

License

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Authors who publish with this journal agree to the following terms:

The copyright to this article is transferred to JTERA if and when the article is accepted for publication. The undersigned hereby transfers any and all rights in and to the paper including without limitation all copyrights to JTERA. The undersigned hereby represents and warrants that the paper is original and that he/she is the author of the paper, except for material that is clearly identified as to its original source, with permission notices from the copyright owners where required. The undersigned represents that he/she has the power and authority to make and execute this assignment. The copyright transfer form will be sent via email before the article published.

The corresponding author signs for and accepts responsibility for releasing this material on behalf of any and all co-authors. This agreement is to be signed by at least one of the authors who have obtained the assent of the co-author(s) where applicable. After submission of this agreement signed by the corresponding author, changes of authorship or in the order of the authors listed will not be accepted.

How to Cite

[1]
“Performance Comparison of YOLOv8n and Faster R-CNN for Automatic Detection of School Uniform Attributes”, JTERA, vol. 11, no. 1, pp. 189–198, Jun. 2026, doi: 10.31544/jtera.v11.i1.2026.189-198.