Please use this identifier to cite or link to this item:
http://repository.ipb.ac.id/handle/123456789/109152| Title: | Identifikasi Keparahan Penyakit Busuk Pangkal Batang pada Perkebunan Kelapa Sawit (Elaeis guineensis Jacq.) Menggunakan Sentinel-2 |
| Other Titles: | Identifying the Severity of Basal Stem Rot Disease in Oil Palm (Elaeis guineensis Jacq.) Plantation Using Sentinel-2 |
| Authors: | Trisasongko, Bambang Panuju, Dyah Handrian, Ramadhani Dwi |
| Issue Date: | 7-Sep-2021 |
| Publisher: | IPB University |
| Abstract: | Kelapa sawit merupakan sektor perkebunan yang paling diunggulkan di
Indonesia. Pemantauan perkebunan skala besar membutuhkan kekhususan dalam
pengumpulan data dan analisisnya. Pengamatan cakupan perkebunan telah lama
dilakukan dengan menggunakan data penginderaan jauh. Dengan berbagai masalah
yang kompleks di perkebunan kelapa sawit, fokus riset umumnya berada dalam
ranah perkiraan nutrisi daun dan untuk mengidentifikasi dampak penyakit tanaman.
Fokus riset terakhir ini perlu menjadi perhatian, terutama pada penyakit busuk
pangkal batang (BPB) yang disebabkan oleh cendawan Ganoderma boninense.
Data multispektral Sentinel-2 yang tersedia secara bebas dimanfaatkan dalam
penelitian ini. Identifikasi gejala awal dampak BPB dianalisis dengan
menggunakan pendekatan pemelajaran mesin yang populer, yaitu random forest
(RF). Data lapangan sebagai referensi dikumpulkan di Kebun Kelapa Sawit
Cikasungka, Kabupaten Bogor. Dengan menggunakan model RF yang disetel
(tuned), akurasi keseluruhan sebesar 82% diperoleh dengan lima target berbeda,
yaitu empat tingkat keparahan dan pohon sehat. Penelitian ini juga menyimpulkan
bahwa data tambahan seperti indeks vegetasi mampu membedakan tingkat
keparahan penyakit meskipun tidak responsif. Penelitian ini menunjukkan bahwa
model RF yang disetel sangat bermanfaat untuk membangun model pemelajaran
mesin yang tepat dan adaptif. Oil palm remains the most prominent plantation sector in Indonesia. Monitoring large scale plantations requires specificity in datasets and approaches. Observing the extent of plantations has long been conducted using remotely-sensed data. With the complex issues in oil palm plantations, focus has also been made in investigating the utility of earth-observing satellites to estimate foliar nutrients and to identify the impact of plant diseases. The latter has been one of research interests, especially the basal stem rot (BSR) caused by Ganoderma boninense. This research proposed a freely-available, multispectral data of Sentinel-2, taking its advantage of frequent observation period. Preliminary identification of BSR impacts was examined by using a popular machine learning approach, i.e. random forests (RF). Ground data as the reference were collected in Cikasungka plantation, Bogor Regency. Using tuned RF model, overall accuracy was about 82% with five distinctive targets, i.e. four severity levels and the normal trees. Additional data, such as the vegetation index, were able to distinguish the severity of the disease, although they were irresponsive. This research suggests that tuned random forest model could be invaluable for constructing a proper machine learning model that adaptive to data feeds. |
| URI: | http://repository.ipb.ac.id/handle/123456789/109152 |
| Appears in Collections: | UT - Soil Science and Land Resources |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| Cover.pdf | Cover | 556.62 kB | Adobe PDF | View/Open |
| Full Text.pdf Restricted Access | Fullteks | 1.9 MB | Adobe PDF | View/Open |
| Lampiran.pdf | Lampiran | 557.49 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.