Implementation of Image Segmentation Using YOLOv8 for Automatic Road Condition Assessment
Implementasi Segmentasi Citra Menggunakan YOLOv8 untuk Penilaian Kondisi Jalan Otomatis
DOI:
https://doi.org/10.21070/ups.9829Keywords:
YOLOv8, image segmentation, road damage, pothole detection, computer visionAbstract
Road infrastructure conditions play a crucial role in supporting transportation safety, comfort, and efficiency. The objective of this research is to develop a system that detects road damage in the form of potholes using pixel segmentation masks to quantitatively calculate the percentage and area of damage to the road surface. This system implements the YOLOv8 model with a deep learning-based image segmentation approach to detect and assess road damage, particularly potholes. A dataset of 780 road images was used as training and test data, with a labeling process based on segmentation masks. The model was trained using the YOLOv8n-seg architecture and evaluated based on performance metrics such as precision, recall, and mean Average Precision (mAP) for 200 epochs on the Kaggle platform, resulting in the best performance with an mAP50(box) score of 0.708 and mAP(mask) of 0.723.
Downloads
References
S. Liao, “Road damage detection algorithm based on optimised You only Look Once version 8,” 2024 5th Int. Conf. Comput. Eng. Appl. ICCEA 2024, pp. 1381–1384, 2024, doi: 10.1109/ICCEA62105.2024.10603714.
Z. Sun, L. Zhu, S. Qin, Y. Yu, R. Ju, and Q. Li, “Road Surface Defect Detection Algorithm Based on YOLOv8,” Electron., vol. 13, no. 12, pp. 1–17, 2024, doi: 10.3390/electronics13122413.
B. P. statistik (BPS), “Statsitik Transportasi Darat,” p. 75, 2021, doi: 06100.2264.
Y. Jiang, “Road damage detection and classification using deep neural networks,” Discov. Appl. Sci., no. July, 2024, doi: 10.1007/s42452-024-06129-0.
K. Zeng, R. Fan, and X. Tang, “Efficient and accurate road crack detection technology based on YOLOv8-ES,” Auton. Intell. Syst., vol. 5, no. 1, 2025, doi: 10.1007/s43684-025-00091-3.
C. Shorten and T. M. Khoshgoftaar, “A survey on Image Data Augmentation for Deep Learning,” J. Big Data, 2019, doi: 10.1186/s40537-019-0197-0.
E. Royhan et al., “PENGEMBANGAN SISTEM DETEKSI LUBANG PADA JALAN MENGGUNAKAN ALGORITMA YOLO BERBASIS ESP32-CAM,” vol. 9, no. 4, pp. 1–12, 2025.
Y. Zanevych, V. Yovbak, O. Basystiuk, N. Shakhovska, and S. Fedushko, “Evaluation of Pothole Detection Performance Using Deep Learning Models Under Low-Light Conditions,” 2024.
M. Yurdakul and Ş. Taşdemİr, “An Enhanced YOLOv8 Model for Real-Time and Accurate Pothole Detection and Measurement”.
Saeeda Varawalla, “Enhancing Road Safety: Real-Time Surface Anomaly Detection Using YOLOv8,” J. Inf. Syst. Eng. Manag., vol. 10, no. 20s, pp. 543–551, 2025, doi: 10.52783/jisem.v10i20s.3177.
R. S. Wijaya, A. Wibisana, and E. R. Jamzuri, “Comparative Study of YOLOv5 , YOLOv7 and YOLOv8 for Robust Outdoor Detection,” 2024.
A. U. Amri and G. P. Kusuma, “Comparative study of pothole detection using deep learning on smartphone,” vol. 37, no. 2, pp. 995–1004, 2025, doi: 10.11591/ijeecs.v37.i2.pp995-1004.
Z. Demirel, S. T. Nasraldeen, Ö. Pehlivan, S. Shoman, and M. Albdairi, “Comparative Evaluation of YOLO and Gemini AI Models for Road Damage Detection and Mapping,” pp. 1–26, 2025.
Farzad, “Pothole Segmentation YOLOv8.” Accessed: Dec. 21, 2025. [Online]. Available: https://universe.roboflow.com/farzad/pothole_segmentation_yolov8/dataset/1
Downloads
Additional Files
Posted
License
Copyright (c) 2026 UMSIDA Preprints Server
This work is licensed under a Creative Commons Attribution 4.0 International License.
