Electronic Theses and Dissertation

Muliadi. 1905108010062. Penggunaan Metode Deep Learning Untuk Deteksi Tanaman Kelapa Sawit Menggunakan ArcGIS Pro (Studi Kasus: Kebun Kelapa Sawit Rakyat Desa Blang Kuta, Kecamatan Peudawa Rayeuk). Dibawah bimbingan Muhammad Rusdi sebagai pembimbing utama dan Sugianto sebagai pembimbing anggota.

RINGKASAN
Perkembangan teknologi Artificial Inteligence (AI) sudah banyak diterapkan dalam proses perkembangan tanaman kelapa sawit, Seperti deteksi tanaman kelapa sawit yang memberikan kemudahan kepada petani untuk perencanaan pemupukan dan monitoring hasil panen. Deteksi tanaman kelapa sawit bisa dilakukan dengan memanfaatkan citra resolusi sangat tinggi (drone) yang dianalisis menggunakan pemodelan dari metode deep learning.
Penelitian ini dilakukan untuk membuat rancang bangun model deep learning untuk proses deteksi secara otomatis tanaman kelapa sawit pada software ArcGIS Pro, serta melihat kinerja dari model deep learning dalam mendeteksi tanaman kelapa sawit. Dalam penelitian ini metode yang digunakan adalah metode analisis modeling.
Hasil deteksi tanaman kelapa sawit menggunakan model deep learning menggunakan citra resolusi sangat tinggi (drone) pada tanaman kelapa sawit diperoleh hasil pelatihan model dengan average precision paling rendah berada pada epoch 0 dengan nilai 0.291454466 dan nilai paling tinggi pada epoch 18 dengan nilai 0.55952686. Hasil deteksi menghasilkan nilai presisi 80,06 %, recall 91,36% dan akurasi 85.71%.

Muliadi. 1905108010062. Utilizing Deep Learning Technique for Oil Palm Plant Detection Using ArcGIS Pro: A Case Study Steaklholder Oil Palm Plantation Is Blang Kuta Village, Peudawa Rayeuk Sub District. Under the guidance of Muhammad Rusdi as the main supervisor and Sugianto as the member mentor. SUMMARY The development of Artificial Intelligence (AI) technology has been widely applied in the process of developing oil palm plants, such as the detection of oil palm plants which makes it easier for farmers to plan fertilization and monitor crop yields. Detection of oil palm plants can be done by utilizing very high-resolution images (drones) analyzed using modeling from deep learning methods. This research was conducted to design a deep learning model for the automatic detection process of oil palm plants in ArcGIS Pro software, as well as to see the performance of deep learning models in detecting oil palm plants. In this study, the method used is the modeling analysis method. The results of detection of oil palm plants using deep learning models using very high resolution images (drones) on oil palm plants obtained model training results with the lowest average precision being at epoch 0 with a value of 0.291454466 and the highest value at epoch 18 with a value of 0.55952686. The detection results resulted in a precision value of 80.06%, recall of 91.36% and accuracy of 85.71%.