Muhammad Ramadhan Maulana. A, - (2023) PEMETAAN DAN IDENTIFIKASI KESIAPAN PETIK TANAMAN TEH BERDASARKAN CITRA DRONE MENGGUNAKAN MASK REGION-BASED CONVOLUTIONAL NEURAL NETWORK (MASK R-CNN) DAN GREEN LEAF INDEX (GLI). S1 thesis, Universitas Pendidikan Indonesia.
![[img]](http://repository.upi.edu/style/images/fileicons/text.png) |
Text
S_KOM_1803771_Title.pdf Download (809kB) |
|
Text
S_KOM_1803771_Chapter1.pdf Download (198kB) |
|
Text
S_KOM_1803771_Chapter2.pdf Restricted to Staf Perpustakaan Download (5MB) |
|
Text
S_KOM_1803771_Chapter3.pdf Download (344kB) |
|
Text
S_KOM_1803771_Chapter4.pdf Restricted to Staf Perpustakaan Download (35MB) |
|
Text
S_KOM_1803771_Chapter5.pdf Download (150kB) |
Abstract
Sebagai negara agraris, sektor pertanian memainkan peran krusial dalam perekonomian Indonesia. Teh, sebagai salah satu komoditi utama ekspor setelah minyak dan gas, menjadi penopang devisa negara. Manajemen perkebunan teh menghadapi tantangan monitoring yang signifikan, terutama dengan karakteristik lahan yang luas dan kompleks. Proses monitoring konvensional dengan melibatkan banyak tenaga pekerja dan waktu yang lama menjadi hambatan utama. Penelitian ini menjelaskan implementasi teknologi drone dalam pemetaan dan identifikasi kesiapan petik tanaman teh. Diawali dengan akuisisi data citra hasil tangkapan drone lalu diolah untuk dijadikan data untuk penggabungan citra dan dataset untuk model pemetaan dan klasifikasi kesiapan petik teh. Selanjutnya implementasi metode image stitching menghasilkan data citra area kebun teh yang luas dari citra tangkapan drone. Kemudian dilakukan pemetaan terhadap area tanaman teh berdasarkan citra hasil gabungan menggunakan metode Mask Region-Based Convolutional Neural Network (Mask R-CNN). Hingga pada akhirnya klasifikasi dilakukan terhadap hasil pemetaan untuk mengidentifikasi tingkat kesiapan petik area tenaman teh tersebut berdasarkan nilai Green Leaf Index (GLI) dan juga Artificial Neural Network, model sequential. Model Mask R-CNN berhasil melakukan pemetaan area tanaman teh dengan rata-rata tingkat presisi sebesar 39%, serta identifikasi tingkat kehijauan teh menggunakan (GLI) serta model sequential mampu mengklasifikasikan kesiapan petik tanaman teh, dengan akurasi mencapai 82% dan nilai loss sebesar 0.4202. Hasil penelitian ini memberikan kontribusi dalam meningkatkan efisiensi monitoring perkebunan teh, memberikan dasar informasi yang lebih baik untuk pengambilan keputusan, dan secara potensial meningkatkan daya saing industri teh Indonesia. As an agrarian nation, the agricultural sector plays a crucial role in Indonesia's economy. Tea, as one of the primary export commodities following oil and gas, serves as a key contributor to the country's foreign exchange reserves. Tea plantation management faces significant monitoring challenges, particularly due to the vast and complex nature of the land. Conventional monitoring processes, involving extensive manual labor and time, pose a major obstacle. This research elucidates the implementation of drone technology for mapping and identifying the readiness for tea harvest. The process begins with acquiring image data captured by drones, which is then processed for image fusion and dataset creation for mapping and classifying tea harvest readiness models. Subsequently, the implementation of the image stitching method generates extensive image data of tea plantation areas from drone captures. The mapping of tea plant areas is then performed based on the fused images using the Mask Region-Based Convolutional Neural Network (Mask R-CNN) method. Ultimately, classification is conducted on the mapping results to identify the harvest readiness level of tea plant areas based on the Green Leaf Index (GLI) and an Artificial Neural Network, sequential model. The Mask R-CNN model successfully maps tea plant areas with an average precision of 39%, while the GLI and sequential model classify tea plant harvest readiness with an accuracy of 82% and a loss value of 0.4202. This research contributes to enhancing the efficiency of tea plantation monitoring, providing a better information foundation for decision-making, and potentially boosting the competitiveness of Indonesia's tea industry.
| Item Type: | Thesis (S1) |
|---|---|
| Additional Information: | https://scholar.google.com/citations?hl=en&view_op=list_works&gmla=AH70aAVknDg-9ORDGupmj4XMJWiaQ9o8ja19-4SaGIB7x0kvrEbD399eqxqrcWttN7OZZgxIRPPhwRQCaG9kJg&user=USMwbZoAAAAJ ID SINTA Dosen Pembimbing: Eddy Prasetyo Nugroho: 5990993 Rosa Ariani Sukamto: 5974496 |
| Uncontrolled Keywords: | citra drone, green leaf index, image stitching, mask region-based convolutional neural network, pemantauan pemetikan teh, pemetaan area tanaman teh drone imagery, green leaf index, image stitching, mask region-based convolutional neural network, tea picking monitoring, tea plant area mapping |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software |
| Divisions: | Fakultas Pendidikan Matematika dan Ilmu Pengetahuan Alam > Program Studi Ilmu Komputer |
| Depositing User: | Muhammad Ramadhan Maulana. A |
| Date Deposited: | 08 Jan 2024 02:55 |
| Last Modified: | 08 Jan 2024 02:55 |
| URI: | http://repository.upi.edu/id/eprint/114130 |
Actions (login required)
 |
View Item |