Proses Segmentasi pada Object Based Imaged Analysis

Mutiara Ramadhani, Muhammad Rusdi, Abubakar Karim

Abstract


Abstrak. Pemetaan jenis mangrove di Kota Langsa menggunakan metode segmentasi OBIA. Penelitian ini dilakukan untuk mengetahui tingkat akurasi pemetaan jenis mangrove. Penelitian dilakukan melalui beberapa tahapan kegiatan yaitu persiapan, pra pengolahan citra, analisis data, segmentasi dan klasifikasi, uji akurasi. Pada penelitian ini proses klasifikasi yang telah selesai dilakukan mendapatkan 3 level kelas klasifikasi berupa vegetasi dan non vegetasi, mangrove dan non mangrove, serta jenis-jenis mangrove yang terdiri dari Rhizopora sp. ,Ceriops sp. , Bruguiera gymnorizha, dan Xylocarpus granatum. Hasil ini diekspor ke dalam bentuk shapefile (.shp) untuk dapat dihitung luas klasifikasi tiap level kelas pada perangkat lunak ArcGIS 10.8. level 1 terdiri atas kelas vegetasi dan non vegetasi seluas 12.533,50 ha dan 8.856,80 ha, (ii) level 2 dari kelas vegetasi terdiri atas kelas mangrove dan non mangrove dengan luasan 4.558,35 ha dan 7.975,15 ha , sedangkan level 3 dari kelas mangrove terdiri dari Rhizopora sp seluas 1.184,55 ha, Ceriops sp seluas 1.159,10 ha, Bruguiera gymnorizha seluas 1.069,68 ha, dan Xylocarpus granatum seluas 1.113.50 ha. pada kelas vegetasi dan non vegetasi adalah 100% dengan nilai kappa 1, kemudian pada kelas mangrove dan non mangrove adalah 99% dengan nilai kappa 0.99, sedangkan pada kelas jenis mangrove sebesar 100% dengan nilai kappa 1, dimana hasil uji akurasi tersebut termasuk ke dalam kelas sangat kuat.

Segmentation Process in Object Based Image Analysis

Abstract. Mapping of mangrove species in Langsa City using the OBIA. This research was conducted to determine the level of accuracy of mapping mangrove species.The research was conducted through several stages of activities, namely preparation, pre-image processing, data analysis, segmentation and classification, accuracy test. In this study, the classification process that has been completed gets 3 levels of classification classes in the form of vegetation and non-vegetation, mangrove and non-mangrove, as well as mangrove species consisting of Rhizopora sp. ,Ceriops sp. , Bruguiera gymnorizha, and Xylocarpus granatum. These results are exported in the form of a shapefile (.shp) to be able to calculate the classification area for each class level in ArcGIS 10.8 software. level 1 consists of vegetation and non-vegetation classes covering an area of 12,533.50 ha and 8,856.80 ha, (ii) level 2 of the vegetation class consists of mangrove and non-mangrove classes with an area of 4,558.35 ha and 7,975.15 ha, while level 3 from the mangrove class consisting of Rhizopora sp covering an area of 1,184.55 ha, Ceriops sp covering an area of 1,159.10 ha, Bruguiera gymnorizha covering an area of 1,069.68 ha, and Xylocarpus granatum covering an area of 1,113.50 ha. in the vegetation and non-vegetation classes it is 100% with a kappa value of 1, then in the mangrove and non-mangrove classes it is 99% with a kappa value of 0.99, while in the mangrove species class it is 100% with a kappa value of 1, where the accuracy test results are included in very strong class.


Keywords


Segmentasi OBIA, lahan mangrove, klasifikasi ; OBIA segmentation, mangrove land, classification

Full Text:

PDF

References


Hardjanto., 2002. Tinjauan perspektif pengelolaan kawasan mangrove dan peluang rehabilitasinya. Institut Pertanian Stiper. Yogyakarta.

LAPAN. 2021. Pedoman Pengolahan Data Penginderaan Jauh SPOT 6. Jakarta: PUSFATJA.

Macnae, W. 1968. A general account of the fauna of the mangrove swamps on Inhaca Island, Monzambique. Journal Ecolpgy. 50: 93- 128.

Pham, T., Yokoya., D. Bui., K. Yoshino., D. Friess. Remote sensing approaches for monitoring mangrove species, structure, and biomass: Opportunities and challenges.Remote Sens.2019,11, 230

Trimble. 2014. eCognition Developer 9.0 User Guide. Munich, Germany: Trimble Documentation.

Xiaoxia, S., Jixian, Z., dan Zhengjun, L., 2004. A Comparison of Object-Oriented and PixelBased Classification Approachs Using Quickbird Imagery.Chinese Academy of Surveiing and Mapping,Beijing, China.




DOI: https://doi.org/10.17969/jimfp.v7i2.19851

Refbacks

  • There are currently no refbacks.


 


JIM Agribisnis|JIM Agroteknologi|JIM Peternakan|JIM Teknologi Hasil Pertanian|JIM Teknik Pertanian|
JIM Ilmu Tanah|JIM Proteksi Tanaman|JIM Kehutanan


E-ISSN: 2614-6053 2615-2878 Statistic Indexing | Citation


Alamat Tim Redaksi:
Fakultas Pertanian,Universitas Syiah Kuala
Jl. Tgk. Hasan Krueng Kalee No. 3, Kopelma Darussalam,
Banda Aceh, 23111, Indonesia.
Email:jimfp@unsyiah.ac.id