Improving YOLO Performance with Advanced Data Augmentation for Soccer Object Detection
DOI:
https://doi.org/10.30871/jaic.v9i6.11256Keywords:
Football, Data Augmentation, Object Detection, YOLOv8mAbstract
This study developed an object detection system for soccer games using the YOLOv8m algorithm with four main classes: player, goalkeeper, referee, and ball. The dataset, consisting of 372 annotated images, exhibited class imbalance, with significantly fewer ball instances compared to players. The basic YOLOv8m architecture was used without internal modifications, but adjustments were made to the output layer and fine-tuning of the pre-trained weights to adapt to the new dataset. Two models were compared: one without and one with advanced augmentation techniques (mosaic, mixup, cutmix). The experimental results showed an increase in mAP@50 from 74.9% to 81.4% in the augmented model, with a statistically significant difference (p < 0.01). However, model performance still decreased under extreme conditions such as high occlusion, rapid movement, and uneven lighting. The combination of data augmentation, output layer adaptation, and fine-tuning proved effective in improving object detection accuracy and provided the basis for the development of a real-time artificial intelligence-based soccer match analysis system.
Downloads
References
[1] X. Li, et al., “Automatic object detection for behavioural research using YOLOv8,” Behavior Research Methods, vol. 56, pp. 2453–2472, 2024.
[2] F. Zhang, et al., “Small Object Detection with YOLOv8 Algorithm Enhanced by MobileViTv3 and Wise-IoU,” in Proc. ICCPR, 2023, pp. 35–40.
[3] Y. Chen, et al., “Small Object Detection in UAV Images Based on YOLOv8n,” Int. J. Computational Intelligence Systems, vol. 17, no. 1, 2024.
[4] M. Naufal, H. A. Azies, and R. M. Brilianto, “Enhanced Brain Tumor Classification Through Gamma Correction in Deep Learning,” Sistemasi: Jurnal Sistem Informasi, vol. 13, no. 6, pp. 2348–2358, Oct. 2024.
[5] A. Ilyana, Nurdin, and Maryana, “Real-Time Detection of Coffee Cherry Ripeness Using YOLOv11,” Journal of Applied Informatics and Computing (JAIC), vol. 9, no. 4, pp. 1170–1178, Aug. 2025
[6] YOLOv8,” J. Computational Theoretical and Applied Mechanics, vol. 1, no. 1, 2024.
[7] J. Liu, et al., “Improved YOLOv8 for Small Object Detection,” in Proc. ICCN/IANet, 2024, pp. 65–72. doi: 10.1145/3670105.3670150.
[8] H. Wang, et al., “Detection of Small Object based on Improved-YOLOv8,” Frontiers in Computing and Intelligent Systems, vol. 3, no. 2, pp. 115–128, 2024.
[9] M. Khan, et al., “An improved YOLOv8 model for prohibited item detection with deformable convolution and dynamic head,” J. Real-Time Image Process., 2025. doi:10.1145/3394171.3413828
[10] L. Zhou, et al., “A Small-Object Detection Model Based on Improved YOLOv8s for UAV Image Scenarios,” Remote Sensing, vol. 16, no. 13, p. 2465, 2024.
[11] R. Pramasetya, et al., “Visual Entity Object Detection System in Soccer Matches Based on Various YOLO Architecture,” Jurnal Teknik Informatika (JUTIF), vol. 5, no. 3, 2024.
[12] L. Yang, et al., “SFG-YOLOv8: efficient and lightweight small-feature gesture keypoint detector,” J. King Saud Univ. – Comput. Inf. Sci., 2025.
[13] T. Ahmed, et al., “Object Detection and Tracking for Football Data Analytics,” in Proc. IACIDS EAI, 2024.
[14] Y. Li, et al., “SOD-YOLO: Small-Object-Detection Algorithm Based on Improved YOLOv8 for UAV Images,” Remote Sensing, vol. 16, no. 16, p. 3057, 2024.
[15] H. Ryu, et al., “YOLOv8 with Post-Processing for Small Object Detection Enhancement,” Applied Sciences, vol. 15, no. 13, p. 7275, 2025.
[16] Muljono, S. A. Wulandari, H. A. Azies, M. Naufal, W. A. Prasetyanto, and F. A. Zahra, “Breaking Boundaries in Diagnosis: Non-Invasive Anemia Detection Empowered by AI,” IEEE Access, vol. 12, pp. 133578–133588, Jan. 2024, doi: 10.1109/ACCESS.2024.3353788.
[17] A. Khalili and B. Smyth, “SOD-YOLOv8—Enhancing YOLOv8 for Small Object Detection in Aerial Imagery and Traffic Scenes,” Sensors, vol. 24, no. 19, p. 6209, 2024.
[18] S. Postupaiev, et al., “Real-Time Camera Operator Segmentation with YOLOv8 in Football Video Broadcasts,” AI, vol. 5, no. 2, pp. 842–872, 2024.
[19] A. S. Fan, YOLOv8-Lite: A Lightweight Object Detection Model for Real-Time Autonomous Driving Systems, TETAI, vol. 1, no. 1, pp. 1-16, Apr. 2024. doi: 10.62762/TETAI.2024.894227
[20] G. Corder and D. Foreman, Nonparametric Statistics: A Step-by-Step Approach, 2nd ed. Hoboken, NJ: Wiley, 2014.
[21] A. Pramasetya et al., Research on Small Object Detection Algorithm Based on YOLOv8, in Proc. CSAI'24, Beijing, Dec. 2024, pp. (paper number) doi :10.1145/3709026.3709027
[22] Y. Zheng et al., HTFD-YOLO: Small Target Detection in Drone Aerial Photography Based on YOLOv8s, J. Supercomputing, vol. 81, art. no. 545, Feb. 2025. doi: 10.1007/s11227-025-07067-3
[23] X. Fan et al., Improvement of YOLOv8 for Vehicle Small Object Detection Research, in Proc. AIAHPC'24, Zhuhai, Jul. 2024.doi: 10.1145/3690931.3690946
[24] Anonymous (2025), Enhanced YOLOv8 for Small-Object Detection in Multiscale UAV Imagery: Innovations in Detection Accuracy and Efficiency, Digital Signal Processing, vol. 158, art. no. 104964, Mar. 2025. doi: 10.1016/j.dsp.2024.104964
[25] Roboflow, Soccer Dataset for Object Detection (Referee, Goalkeeper, Player,Ball).[Online].Available: https://universe.roboflow.com/roboflow-jvuqo/football-players-detection-3zvbc
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Rahayuning Febriyanti Puspita, Muhammad Naufal, Farrikh Al Zami
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) ) that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
