Please use this identifier to cite or link to this item: http://repository.ipb.ac.id/handle/123456789/169278
Title: Pemetaan Habitat Bentik Menggunakan Drone dengan Metode Object Based Image Analysis di Pulau Sebesi, Lampung Selatan.
Other Titles: 
Authors: Agus, Syamsul Bahri
Susilo, Setyo Budi
Nurdin, Wijdan Rafi
Issue Date: 2025
Publisher: IPB University
Abstract: Habitat bentik di memiliki fungsi vital dalam menjaga keseimbangan ekosistem pesisir, sehingga upaya diperlukan upaya untuk memantau dan menjaganya. Namun, keterbatasan metode survei konvensional, terutama mempertimbangkan luasnya daerah pesisir di Indonesia sering kali menjadi hambatan dalam melakukan pemantauan yang cepat dan akurat. Penelitian ini bertujuan untuk memetakan tutupan substrat bentik di pesisir timur Pulau Sebesi, Lampung Selatan, dengan memanfaatkan citra udara dari drone serta pendekatan Object Based Image Analysis (OBIA). Pengambilan data dilakukan menggunakan drone DJI Mavic 2 Pro, kemudian diproses melalui perangkat lunak Agisoft Metashape, eCognition Developer. Proses klasifikasi yang dilakukan dalam dua level mengklasifikasi tujuh tipe substrat bentik berdasarkan algoritma Support Vector Machine (SVM). Hasil klasifikasi menunjukkan bahwa pasir merupakan tutupan paling luas (36,07%), disusul oleh karang keras (33,06%) dan lumpur (13,31%). Nilai akurasi keseluruhan mencapai 82,68%.
Benthic habitat plays a vital role in maintaining balancer in coastal ecosystems, making it important to monitor and protect them. However, the limitations of conventional survey methods espescially due to how vast the coastal area is in Indonesia, its often challenging to do rapid and accurate monitoring. This study aims to map benthic substrate cover along the eastern coast of Sebesi Island, South Lampung, using drone imagery and Object Based Image Analysis (OBIA) approach. Data were collected using a DJI Mavic 2 Pro drone and processed using Agisoft Metashape and eCognition Developer software. The classification process, carried out in two levels, categorized seven types of benthic substrates using the Support Vector Machine (SVM) algorithm. The classification results showed that sand was the most dominant class (36.07%), followed by hard coral (33.06%) and mud (12.94%). The overall classification accuracy reached 82.68%.
URI: http://repository.ipb.ac.id/handle/123456789/169278
Appears in Collections:UT - Marine Science And Technology

Files in This Item:
File Description SizeFormat 
cover_C54180043_cfa88684b1e84d339aead53ece2fd047.pdfCover1.67 MBAdobe PDFView/Open
fulltext_C54180043_91b0b560bd754f8dbfe6c0d6f911d899.pdf
  Restricted Access
Fulltext4.12 MBAdobe PDFView/Open
lampiran_C54180043_e837030349d1459eb3f30ddfef628dac.pdf
  Restricted Access
Lampiran1.56 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.