View Item 
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Forestry and Environment
      • UT - Forest Management
      • View Item
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Forestry and Environment
      • UT - Forest Management
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Pengembangan Algoritma Pendeteksi Devegetasi dan Pertumbuhan Vegetasi Berbasis Analisis Vektor Perubahan dan Pembelajaran Mesin

      Thumbnail
      View/Open
      Cover (1.594Mb)
      Fulltext (5.431Mb)
      Lampiran (408.4Kb)
      Date
      2024
      Author
      Indra, Figo Valentino
      Jaya, I Nengah Surati
      Metadata
      Show full item record
      Abstract
      Tulisan ini menjelaskan pengembangan algoritma untuk mendeteksi devegetasi dan pertumbuhan vegetasi dengan pendekatan pembelajaran mesin, metode pohon keputusan. Algoritma yang dikembangkan dalam penelitian ini menggunakan variabel yang dibangun menggunakan analisis vektor perubahan (Change Vector Analysis/CVA), yaitu variabel magnitude dan direction. Tujuan utama penelitian ini adalah menemukan algoritma pohon keputusan dan variabel terbaik untuk mendeteksi devegetasi dan pertumbuhan vegetasi. Penelitian ini dilakukan di Kabupaten Bandung, dengan data utama berupa data Landsat TM multi-waktu, rekaman tahun 2007 dan 2022 dan data penunjang biogeofisik. Studi ini menemukan bahwa algoritma deteksi devegetasi dan pertumbuhan vegetasi menggunakan kombinasi peubah CVA (magnitude dan direction), dan variabel elevasi menghasilkan akurasi umum (Overall Accuracy/OA) sebesar 98.2% dan akurasi Kappa (Kappa Accuracy/KA) sebesar 95.3%. Penggunaan peubah CVA saja menghasilkan akurasi yang lebih rendah yaitu 92.6% akurasi umum dan 79.9% akurasi Kappa. Studi ini menyimpulkan bahwa algoritma terbaik diperoleh menggunakan pohon keputusan pembelajar mesin dengan parameter information gain dan variabel gabungan CVA (magnitude dan direction) dan elevasi.
       
      This paper describes the development of an algorithm to detect changes in forest and land cover using decision tree of machine learning approach. The algorithm was developed using variables derived from change vector analysis (CVA), namely magnitude and direction. The main objective of this research is to find the best decision tree algorithm and variables for detecting devegetation and regrowth. The study was performed in Bandung Regency, supported using multi-temporal Landsat TM data, recorded in 2007 and 2022 and biophysical data. This study found that the algorithm with CVA (magnitude and direction) and elevation, provided Overall Accuracy of 98.2% and Kappa Accuracy of 95.3%. The use of CVA variable alone provided lesser accuracy of only 92.6% for OA and 79.9% for KA. This study concludes that the best algorithm was obtained using decision tree of machine learning with information gain parameter and the combination of CVA and elevation.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/158069
      Collections
      • UT - Forest Management [3207]

      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