Electronic Theses and Dissertation

Perkembangan teknologi penginderaan jauh di sektor perkebunan mendorong
pemanfaatan UAV/drone sebagai alat pemantauan tanaman yang lebih cepat dan efisien
dibandingkan metode konvensional. Pada perkebunan kelapa sawit, tinggi tanaman
merupakan parameter penting yang berkaitan dengan pertumbuhan, efisiensi panen, dan
evaluasi keseragaman blok. Namun, pengukuran manual pada skala kebun yang luas
cenderung memerlukan waktu dan tenaga besar, serta rentan terhadap human error.
Penelitian ini merumuskan permasalahan mengenai bagaimana tingkat akurasi
pengukuran tinggi tanaman kelapa sawit menggunakan drone dibandingkan metode
konvensional di lapangan. Tujuannya adalah mengetahui serta mengevaluasi tingkat akurasi
pengukuran tinggi tanaman kelapa sawit menggunakan drone dengan membandingkannya
terhadap pengukuran konvensional.
Penelitian dilaksanakan pada kebun kelapa sawit milik warga di Desa Lueng Gayo,
Kecamatan Teunom, Kabupaten Aceh Jaya, sedangkan analisis data dilakukan di
Laboratorium Penginderaan Jauh dan Kartografi Departemen Ilmu Tanah, Fakultas
Pertanian Universitas Syiah Kuala. Pelaksanaan penelitian berlangsung mulai Juli 2025
sampai dengan selesai. Metode yang digunakan adalah deskriptif kuantitatif, dengan
estimasi tinggi tanaman berbasis CHM yang diperoleh dari selisih Digital Surface Model
(DSM) dan Digital Terrain Model (DTM). Data dikumpulkan melalui akuisisi citra drone
dan pengolahan fotogrametri, kemudian dilakukan validasi lapangan menggunakan
hagameter serta uji akurasi menggunakan Root Mean Square Error (RMSE).
Hasil pengolahan CHM menunjukkan tinggi tanaman kelapa sawit bervariasi antara
5,7 m hingga 9,0 m dengan rata-rata 7,12 m. Berdasarkan perbandingan dengan pengukuran
hagameter, diperoleh RMSE sebesar 1,33 m dengan selisih rata-rata 1,76 m. Secara umum,
estimasi tinggi tanaman menggunakan UAV-CHM menunjukkan akurasi yang cukup baik
untuk pemantauan kebun dan evaluasi pertumbuhan pada skala blok, sehingga dapat
dipertimbangkan sebagai alternatif yang lebih efisien dibanding pengukuran manual.

The development of remote sensing technology in the plantation sector has encouraged the use of Unmanned Aerial Vehicles (UAVs) or drones as a faster and more efficient tool for crop monitoring compared to conventional methods. In oil palm plantations, plant height is an important parameter related to growth assessment, harvesting efficiency, and block uniformity evaluation. However, manual measurements at a large plantation scale tend to require considerable time and labor and are prone to human error. This study addresses the issue of the accuracy level of oil palm height measurement using UAVs compared to conventional field measurement methods. The objective of this research is to determine and evaluate the accuracy of oil palm height estimation using UAVs by comparing it with conventional measurements. The research was conducted on a community-owned oil palm plantation in Lueng Gayo Village, Teunom Subdistrict, Aceh Jaya Regency, while data analysis was carried out at the Remote Sensing and Cartography Laboratory, Department of Soil Science, Faculty of Agriculture, Syiah Kuala University. The study was conducted from July 2025 until completion. A quantitative descriptive method was applied, with plant height estimation based on the Canopy Height Model (CHM), obtained from the difference between the Digital Surface Model (DSM) and the Digital Terrain Model (DTM). Data were collected through UAV image acquisition and photogrammetric processing, followed by field validation using a hagameter and accuracy assessment using the Root Mean Square Error (RMSE). The CHM processing results showed that oil palm heights ranged from 5.7 m to 9.0 m, with an average height of 7.12 m. Based on the comparison with hagameter measurements, an RMSE value of 1.33 m and an average difference of 1.76 m were obtained. Overall, oil palm height estimation using the UAV-CHM method demonstrated a reasonably good level of accuracy for plantation monitoring and growth evaluation at the block scale, indicating that this method can be considered a more efficient alternative to manual measurement.