PENGOLAHAN DATA CITRA PADA SISTEM DETEKSI DAN MAPPING JALAN BERLUBANG MENGGUNAKAN DEEP LEARNING DAN GPS DI WILAYAH SUMATERA SELATAN

HUMAIROH, HILDIANA and Irmawan, Irmawan (2024) PENGOLAHAN DATA CITRA PADA SISTEM DETEKSI DAN MAPPING JALAN BERLUBANG MENGGUNAKAN DEEP LEARNING DAN GPS DI WILAYAH SUMATERA SELATAN. Undergraduate thesis, Sriwijaya University.

This is the latest version of this item.

[thumbnail of RAMA_20201_03041182025018.pdf] Text
RAMA_20201_03041182025018.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (47MB) | Request a copy
[thumbnail of RAMA_20201_03041182025018_TURNITIN.pdf] Text
RAMA_20201_03041182025018_TURNITIN.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (17MB) | Request a copy
[thumbnail of RAMA_20201_03041182025018_0004045903_01_front_ref.pdf] Text
RAMA_20201_03041182025018_0004045903_01_front_ref.pdf - Accepted Version
Available under License Creative Commons Public Domain Dedication.

Download (1MB)
[thumbnail of RAMA_20201_03041182025018_0004045903_02.pdf] Text
RAMA_20201_03041182025018_0004045903_02.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (1MB) | Request a copy
[thumbnail of RAMA_20201_03041182025018_0004045903_03.pdf] Text
RAMA_20201_03041182025018_0004045903_03.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (1MB) | Request a copy
[thumbnail of RAMA_20201_03041182025018_0004045903_04.pdf] Text
RAMA_20201_03041182025018_0004045903_04.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (3MB) | Request a copy
[thumbnail of RAMA_20201_03041182025018_0004045903_05.pdf] Text
RAMA_20201_03041182025018_0004045903_05.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (407kB) | Request a copy
[thumbnail of RAMA_20201_03041182025018_0004045903_06_ref.pdf] Text
RAMA_20201_03041182025018_0004045903_06_ref.pdf - Bibliography
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (331kB) | Request a copy
[thumbnail of RAMA_20201_03041182025018_0004045903_07_lamp.pdf] Text
RAMA_20201_03041182025018_0004045903_07_lamp.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (4MB) | Request a copy

Abstract

According to data from the Ministry of Public Works and Public Housing (PUPR) in 2022, South Sumatra is one of the regions with a high rate of road damage in Indonesia, including numerous potholes. To address this issue, an automatic pothole detection system using deep learning and GPS-based mapping was developed. This system involves data preprocessing and mapping to create a representative dataset for training the deep learning model to recognize the patterns and characteristics of potholes. The process includes data mining and image processing techniques such as object detection, segmentation, and pattern recognition. This study focuses on collecting datasets in South Sumatra. Data collection will be conducted using a ground vehicle equipped with a GoPro Hero 8 action camera for real-time video capture and GPS for pothole mapping. The collected data must accurately reflect pothole conditions, with varying camera angles to ensure precise representation. The collected data is used to train the deep learning model to identify various patterns and characteristics of potholes. This research successfully implemented image data processing in a pothole detection and mapping system using deep learning and GPS, achieving average error UIQI values 0,074, indicating high similarity images and GPS Neo-6M demonstrated good accuracy with an average distance error below 1 meter. Data preprocessing significantly improved the performance of YOLOv8 and MaskRCNN models, proving that preprocessing helps the model achieve better performance across various evaluation metrics.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Deep Learning, GPS, Preprocessing Data, Sumatera Selatan.
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK1-9971 Electrical engineering. Electronics. Nuclear engineering > TK1 Electrical engineering--Periodicals. Automatic control--Periodicals. Computer science--Periodicals. Information technology--Periodicals. Automatic control. Computer science. Electrical engineering. Information technology.
Divisions: 03-Faculty of Engineering > 20201-Electrical Engineering (S1)
Depositing User: Hildiana Humairoh
Date Deposited: 18 Jul 2024 03:17
Last Modified: 18 Jul 2024 03:17
URI: http://repository.unsri.ac.id/id/eprint/151571

Available Versions of this Item

Actions (login required)

View Item View Item