Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Inovate : Jurnal Ilmiah Inovasi Teknologi Informasi
Rosyidah, Aini Nur
Unknown Affiliation
Computer Science & IT

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
DETEKSI KERUSAKAN JALAN MENGGUNAKAN MODEL DEEP LEARNING YOLOv8 Agus Prastyo, Edwin Hari; Rosyidah, Aini Nur
Inovate Vol 10 No 1 (2025): September
Publisher : Fakultas Teknologi Informasi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33752/inovate.v10i1.11198

Abstract

Road damage such as cracks, potholes, and other surface damage are infrastructure problems that require regular detection and maintenance. Manual inspections are time-consuming, costly, and inefficient for extensive road networks. This study implements the YOLOv8 (You Only Look Once version 8) deep learning model for automatic and real-time road damage detection. YOLOv8 was chosen for its advantages in inference speed and high object detection accuracy. The research methodology included collecting a road damage dataset, data preprocessing, data augmentation, training the YOLOv8 model, and evaluating performance using the metrics of precision, recall, F1-score, and mean Average Precision (mAP). The dataset used was RDD2022 (Road Damage Dataset 2022), which contains 26,620 images with bounding box annotations for various types of road damage. The results of the experiment show that the YOLOv8 model achieved a detection accuracy of 86.4%, precision of 67.3%, recall of 65.2%, F1-score of 0.61, and mean Average Precision (mAP) of 62%. The YOLOv8 model also demonstrated an inference speed of 0.5 ms with 0.41 GB of memory usage, proving its efficiency for real-time implementation on mobile devices and vehicle-based inspection systems. Comparison with other methods such as Faster R-CNN and SSD shows that YOLOv8 provides the best trade-off between accuracy and speed. This research contributes to the development of automated road inspection systems that can assist governments and infrastructure managers in more efficient and timely road maintenance.
Co-Authors Agus Prastyo, Edwin Hari