View Item 
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Agriculture
      • UT - Soil Science and Land Resources
      • View Item
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Agriculture
      • UT - Soil Science and Land Resources
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Klasifikasi Tutupan Lahan Kawasan Rural Menggunakan Citra Sentinel-2 Dengan Metode Random Forests

      Thumbnail
      View/Open
      Full text (2.293Mb)
      Cover (1.103Mb)
      Lampiran (325.1Kb)
      Date
      2023-08-22
      Author
      Muharram, M Fadil
      Trisasongko, Bambang Hendro
      Panuju, Dyah Retno
      Metadata
      Show full item record
      Abstract
      Pembangunan wilayah dan pengelolaan sumberdaya memerlukan data tutupan lahan. Dewasa ini, pembangunan tidak hanya berfokus pada kawasan perkotaan melainkan juga sudah mulai merambah ke kawasan perdesaan dan pinggiran kota. Oleh karena itu, pemantauan dan pemetaan tutupan lahan perdesaan (rural) penting untuk dilakukan. Penelitian ini bertujuan untuk mengkaji karakteristik spektral beberapa tutupan lahan dan melakukan pemodelan tutupan lahan rural dengan metode Random Forests (RF). Penelitian dilaksanakan di perbatasan Kabupaten Pandeglang dan Kabupaten Lebak. Percobaan dilakukan dengan menggunakan data citra Sentinel-2 yang memiliki resolusi tinggi. Metode RF digunakan dalam melakukan klasifikasi berbagai jenis tutupan lahan yang diamati yaitu; lahan terbangun, lahan pertanian semusim, kebun campuran, kebun kelapa sawit, kebun karet monokultur, karet rakyat (jungle rubber), dan tanaman kayu (timber). Berdasarkan percobaan yang telah dilakukan diketahui bahwa kombinasi kanal SWIR (Shortwave Infra Red) cukup mampu memisahkan tutupan lahan yang ada dibandingkan kombinasi kanal lainnya. Selain itu, metode RF cukup efektif untuk klasifikasi tutupan lahan meskipun variasi kepercayaan informasi pada tingkat kelas tutupan lahan masih belum sepenuhnya seragam. Penelitian ini menunjukkan bahwa tidak semua kanal yang digunakan pada citra Sentinel-2 berkontribusi terhadap ekstraksi informasi, khususnya pada tutupan lahan yang dinilai cukup sulit dibedakan seperti karet rakyat. Namun demikian, metode klasifikasi RF yang digunakan cukup efektif dengan akurasi total sebesar 77.7%.
       
      Regional development and the management of natural resources require land cover data. The development, nowadays, does not merely set its focus to urban areas, but has been expanded by incorporating rural and suburban areas. Monitoring and mapping of rural land cover is therefore important to accommodate this need. This study aimed to assess spectral characteristics of selected land cover types and to model rural land cover using the Random Forests (RF). The research was conducted over an area bordering Pandeglang and Lebak Regencies. The experiment demonstrated the value of using Sentinel-2 image, which is considerably a high resolution data. The RF method was used in classifying various types of land cover, namely; built-up, seasonal agricultural crops, mixed gardens, oil palm plantations, monoculture rubber plantations, jungle rubber, and timber plants. Based on the experiment, it was discovered that the SWIR (Shortwave Infra Red) channel was quite capable of separating the existing land cover compared to other band combinations. In addition, the RF method is fairly effective for land cover classification although confidence at class level varied. This study showed that not all channels used in Sentinel-2 imagery contributed to the extraction of information, especially on land cover that was considered quite difficult to distinguish such as smallholder rubber. However, the RF classification method was somewhat effective with a total accuracy of 77.7%.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/124180
      Collections
      • UT - Soil Science and Land Resources [2825]

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository
        

       

      Browse

      All of IPB RepositoryCollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

      My Account

      Login

      Application

      google store

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository