Skripsi

SISTEM DETEKSI KERUSAKAN JALAN SECARA REAL- TIME MENGGUNAKAN ALGORITMA DEEP- LEARNING DAN GPS DI WILAYAH KOTA PALEMBANG

---

Road damage is an infrastructure problem that affects road user safety and transportation efficiency. Manual road damage detection is considered inefficient due to the large amount of time, labor, and cost required, as well as its subjective nature. Therefore, this study aims to develop an automatic road damage detection system based on computer vision using deep learning methods. This research applies several algorithmic approaches, namely YOLOv11 Object Detection, YOLOv11 Segmentation, and YOLOv11 Classification, with EfficientDet used as a comparison model. The dataset was obtained from road surface images and videos acquired using a road damage detection robot, followed by preprocessing, annotation, and model training stages. System performance was evaluated based on detection results, detection success rate, and real-time processing capability. The results show that the YOLOv11 Segmentation approach achieved the best performance among the evaluated methods, while the EfficientDet model demonstrated suboptimal performance and is therefore not recommended for implementation in the proposed system.

---

Availability

Detail Information

Series Title

-

Call Number

T1951342026

Publisher

Indralaya : Prodi Teknik Elektro, Fakultas Teknik Universitas Sriwijaya., 2026

Collation

viii, 99 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

910.285 07

Content Type

Text

Media Type

unmediated

Carrier Type

-

Edition

-

Subject(s)

 Prodi Teknik Elektro
Sistem Penentuan Posisi Global (GPS)

Specific Detail Info

-

Statement of Responsibility

KA

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment