dc.contributor.advisor | Jaya, I Nengah Surati | |
dc.contributor.author | Pertiwi, Ni Luh Ayu Anjani Githa | |
dc.date.accessioned | 2021-12-20T14:38:10Z | |
dc.date.available | 2021-12-20T14:38:10Z | |
dc.date.issued | 2021-12-20 | |
dc.identifier.uri | http://repository.ipb.ac.id/handle/123456789/110253 | |
dc.description.abstract | Penelitian ini mengkaji teknik klasifikasi berbasis objek (OBIA) dengan
menggunakan citra resolusi tinggi dan sangat tinggi dalam mendeteksi penutupan
tajuk hutan pada hutan lahan kering dalam membedakan tajuk dan gap di PT. Bina
Balantak Raya, Sulawesi Selatan. Segmentasi yang dilakukan menggunakan
algoritma Watershed, namun algoritma Watershed mempunyai kelemahan yaitu
adanya over segmentation sehingga diperlukan proses filtering yaitu dengan Mean
Filter pada citra SPOT 6 dan citra Worldview 2 sebelum proses segmentasi. Hasil
uji akurasi didapatkan bahwa segmentasi dengan citra SPOT 6 lebih baik
menggunakan filter rata-rata dengan kombinasi parameter (10-0,5,0,5) dengan nilai
OA sebesar 88,1% dan nilai KAsebesar 59,8%, sedangkan untuk hasil segmentasi
citra Worldview 2 lebih baik tanpa filter rata-rata dengan nilai akurasi OA 84,5%
dan KA sebesar 60,43% dengan kombinasi parameter (15-0,4,0,6). | id |
dc.description.abstract | This study examines object-based classification (OBIA) techniques using
high and very high resolution imagery in detecting dry forest canopy in
distinguishing canopy and gap in PT. Bina Balantak Raya, South Sulawesi.
Segmentation is carried out using the Watershed algorithm, but the Watershed
algorithm has a weakness, namely the existence of segmentation so that a filtering
process is needed, namely the Mean Filter on the SPOT 6 image and the Worldview
2 image before the segmentation process. The results of the accuracy test show that
segmentation with SPOT 6 images is better using an average filter with a
combination of parameters (10-0.5,0.5) with an OA value of 88.1% and a KA value
of 59.8%, while for The results of Worldview 2 image segmentation are more
unfiltered with an average accuracy of OA of 84.5% and KA of 60.43% with a
combination of parameters (15-0.4,0.6). | id |
dc.language.iso | id | id |
dc.publisher | IPB University | id |
dc.title | Segmentasi Penutupan Tajuk Hutan Lahan Kering Menggunakan Citra Resolusi Tinggi dan Sangat Tinggi | id |
dc.title.alternative | Segmentation of Dry Land Forest Crown Closure using High Resolution Imagery and Very High Resolution Imagery | id |
dc.type | Undergraduate Thesis | id |
dc.subject.keyword | Crown | id |
dc.subject.keyword | Kappa Accuracy | id |
dc.subject.keyword | Mean filter | id |
dc.subject.keyword | Segmentation | id |
dc.subject.keyword | Watershed | id |