Abyanta, Raihan Daffa Gusti (2023) Analisis Tree Counting Kelapa Sawit Menggunakan Metode Object-Based Image Analysis (OBIA) dan Random Forest Machine Learning (Studi Kasus: Kecamatan Cempaga Hulu, Kabupaten Kotawaringin Timur). Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 03311940000043_Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
03311940000043_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (9MB) | Request a copy |
Abstract
Kelapa sawit merupakan salah satu penghasil minyak nabati terbesar di Indonesia. Potensi yang ada mengakibatkan terbantunya sektor ekonomi dalam bidang perkebunan. Namun pada tahun 2021 volume ekspor minyak sawit di Indonesia mencapai 26.9 juta ton terjadi penurunan yang diakibatkan penyebaran covid-19, sehingga menutup akses ekspor sementara. Hal itu membuat pasokan menurun dikarenakan besarnya perminataan dan minimnya ketersediaan SDM yang ada, sehingga dibutuhkan otomatisasi agar mempermudah proses dan mempercepat monitoring dalam melakukan produksi sawit. Sebagai upaya mempercepat perhitungan jumlah pohon kelapa sawit guna asseting dalam produksi komoditas sawit, dapat menggunakan teknologi di bidang geomatika dengan mengaplikasikan pemanfaatan teknologi machine learning. Salah satu metode yang dapat digunakan untuk perhitungan jumlah pohon kelapa sawit dengan tepat, dan cepat adalah dengan klasifikasi berbasis objek menggunakan data mozaik orthophoto. Data tersebut dapat diolah menggunakan metode klasifikasi yaitu OBIA (Object-Based Image Analysis) dan Random Forest Machine Learning. Dalam pengolahan menggunakan masing masing sample pada sawit, tanah dan non sawit diwakili 80 sampel. Didapatkan hasil berupa raster lalu diolah menggunakan metode multiresolution. Hasil dari pengolahan otomatisasi sebesar 3098 titik pohon sawit dan hasil uji validasi digitasi manual sebesar 3034 titik pohon sawit sehingga didapat deviasi sebesar 64 titik pohon. Melalui pengujian performa pemodelan machine learning, yakni confusion matrix dengan nilai sebesar 0,89 dan kappa sebesar 0,78.
===============================================================================================================================
Palm oil is one of the largest producers of vegetable oil in Indonesia. The potential resulted in helping the economic sector in the plantation sector. However, in 2021 the volume of palm oil exports in Indonesia reached 26.9 million tons, a decrease due to the spread of Covid-19, thereby temporarily closing export access. This causes the supply to decrease due to the large demand and the lack of availability of existing human resources, so automation is needed to simplify the process and speed up monitoring in carrying out palm oil production. To speed up the calculation of the number of oil palm trees for asseting in the production of palm oil commodities, you can use machine learning technology in the field of geomatics. One method that can be used to calculate the number of oil palm trees accurately and quickly is object-based classification using orthophoto mosaic data. The data can be processed using classification methods in the form of Object-Based Image Analysis (OBIA) and Random Forest Machine Learning. In processing based on objects with oil palm, soil, and non-palm oil represented by 80 samples. The results were obtained in the form of a raster and then processed using the multiresolution method. The results obtained from automation processing were 3098 palm tree points and the manual digitization validation test results were 3034 palm tree points, so a deviation of 64 tree points was obtained. Through testing the performance of machine learning modeling, namely the confusion matrix with a value of 0,89 and a kappa of 0,78.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Kelapa Sawit, Tree Counting, Random Forest, OBIA, Machine Learning, Palm Oil |
| Subjects: | G Geography. Anthropology. Recreation > G Geography (General) G Geography. Anthropology. Recreation > G Geography (General) > G70.212 ArcGIS. Geographic information systems. G Geography. Anthropology. Recreation > G Geography (General) > G70.5.I4 Remote sensing |
| Divisions: | Faculty of Civil, Planning, and Geo Engineering (CIVPLAN) > Geomatics Engineering > 29202-(S1) Undergraduate Thesis |
| Depositing User: | Raihan Daffa Gusti Abyanta |
| Date Deposited: | 16 Aug 2023 03:20 |
| Last Modified: | 16 Aug 2023 03:20 |
| URI: | http://repository.its.ac.id/id/eprint/101309 |
Actions (login required)
 |
View Item |