Integration of BDCN-UNet and OBIA for Multisensor Coastline Extraction on the Island of Java

Anastasya, Talitha Atha (2026) Integration of BDCN-UNet and OBIA for Multisensor Coastline Extraction on the Island of Java. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 22081010054.-cover.pdf Download (1MB)
	Text (Bab 1) 22081010054.-bab1.pdf Download (83kB)
	Text (Bab 2) 22081010054.-bab2.pdf Restricted to Repository staff only until 15 June 2029. Download (409kB)
	Text (Bab 3) 22081010054.-bab3.pdf Restricted to Repository staff only until 15 June 2029. Download (2MB)
	Text (Bab 4) 22081010054.-bab4.pdf Restricted to Repository staff only until 15 June 2029. Download (4MB)
	Text (Bab 5) 22081010054.-bab5.pdf Download (72kB)
	Text (Daftar pustaka) 22081010054.-daftarpustaka.pdf Download (95kB)

Abstract

As an archipelago, Indonesia has a long and dynamic coastline, making it vulnerable to changes caused by both natural factors and human activities. The diversity of coastal characteristics, particularly on the island of Java, renders conventional monitoring methods ineffective, necessitating a more adaptive approach. This study utilizes multisensor remote sensing data from Sentinel-1 (SAR) and Sentinel-2 (optical) across several coastal regions of Java. The research stages include preprocessing, reprojection, normalization, data fusion, image patching, labeling, and the division of training and test data. The method used is BDCN-UNet, with a testing scenario comparing the UNet model, BDCN-UNet, and BDCN-UNet with OBIA as post-processing on SAR, optical, and multisensor data. The results show that BDCN-UNet with OBIA delivers the best performance, with an average improvement of 0.3–0.5% in F1-score and IoU, and a reduction in RMSE of 0.2–0.6 m compared to without OBIA. The best performance was achieved on multisensor (Fusion) data with an F1-score of 93.48%, an IoU of 96.35%, and an RMSE of 5.15 m, followed by Sentinel-1 and Sentinel-2 data. These results indicate that the integration of OBIA improves segmentation quality by enhancing the connectivity of the coastline, while the use of multisensor data yields the most optimal results in coastline extraction.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Muttaqin, Faisal	NIDN0030058602	faisalmuttaqin.if@upnjatim.ac.id
Thesis advisor	Anggraeny, Fetty Tri	NIDN0711028201	fettyanggraeny.if@upnjatim.ac.id

Subjects:

Q Science > QA Mathematics > QA76.87 Neural computers

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Talitha Atha Anastasya

Date Deposited:

19 Jun 2026 07:04

Last Modified:

19 Jun 2026 07:36

URI:

https://repository.upnjatim.ac.id/id/eprint/54008