Please use this identifier to cite or link to this item: http://repository.ipb.ac.id/handle/123456789/107241
Title: Penerapan Algoritme Maximum Likelihood dan k-Nearest Neighbour untuk Identifikasi Lahan Bawang Putih Menggunakan Citra Sentinel-1A
Authors: Agmalaro, Muhammad Asyhar
Sitanggang, Imas Sukaesih
Rahmani, Intan Aida
Issue Date: 2021
Publisher: IPB University
Abstract: Komoditas bawang putih (Allium sativum) di Indonesia dikenal memiliki banyak khasiat. Seiring dengan pertumbuhan penduduk, konsumsi bawang putih juga semakin meningkat. Namun, ketersediaannya belum mencukupi seluruh permintaan yang ada. Kementerian Pertanian merencanakan swasembada bawang putih pada 2030 di salah satu sentra penanaman bawang putih Kecamatan Sembalun, Lombok Timur untuk memenuhi permintaan dengan meningkatkan lahan perkebunan. Sehingga penelitian lanjutan untuk identifikasi lahan bawang putih menjadi penting. Penelitian ini bertujuan untuk menerapkan algoritme yang dapat mengindentifikasi lahan bawang putih menggunakan citra Sentinel-1A di Kecamatan Sembalun, Lombok Timur. Model klasifikasi dibuat menggunakan algoritme k-Nearest Neighbour (kNN) dan Maximum Likelihood Classification (MLC). Model klasifikasi memiliki akurasi terbaik yaitu 78,81% pada algoritme kNN dengan koefisien kappa 0,58 dan 76,31% pada algoritme MLC dengan koefisien kappa 0,53. Berdasarkan hasil tersebut, penelitian ini menyimpulkan bahwa mode klasifikasi tersebut dapat mengidentifikasi lahan bawang putih.
The commodity of garlic (Allium sativum) in Indonesia is known to have many health benefits. Along with population growth, the consumption of garlic is increasing. However, its availability is not sufficient for all existing demands. In 2030, The Ministry of Agriculture plans to achieve garlic self-sufficiency in one of the centre garlic field in Sembalun District, East Lombok by enlarging the area. Therefore, further research to identify garlic field is important. This study aims to apply classification algorithms that can identify garlic fields using Sentinel-1A imageries in Sembalun District, East Lombok. The classification model is created using the k-Nearest Neighbour (kNN) algorithm and Maximum Likelihood Classification (MLC). The classification models have the best accuracy 78,81% for the kNN algorithm with a kappa coefficient of 0,58 and 76,31% for the MLC algorithm with a kappa coefficient of 0,53. Based on these results, this study concludes that the classification model is able to identify garlic fields.
URI: http://repository.ipb.ac.id/handle/123456789/107241
Appears in Collections:UT - Computer Science

Files in This Item:
File Description SizeFormat 
Cover, Lembar Pengesahan, Prakata, Daftar Isi.pdf
  Restricted Access
Cover557.19 kBAdobe PDFView/Open
G64170091_Intan Aida Rahmani.pdf
  Restricted Access
Fulltext1.79 MBAdobe PDFView/Open
Lampiran.pdf
  Restricted Access
Lampiran339.84 kBAdobe PDFView/Open


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