Estimasi Jumlah Buah Kelapa Sawit untuk Mendukung Manajemen Produksi Berbasis YOLOv5 TensorFlow Lite

Isi Artikel Utama

Muhammad Bintang
Rasyidah
Novi

Abstrak

Penelitian ini bertujuan untuk mengembangkan aplikasi mobile yang mampu mendeteksi dan menghitung jumlah tandan buah kelapa sawit secara otomatis menggunakan metode deteksi objek YOLOv5. Dataset yang digunakan dalam penelitian ini terdiri dari 226 citra tandan kelapa sawit yang diperoleh langsung dari perkebunan PT Eagle High Plantation di Kalimantan Timur. Dataset tersebut melalui proses anotasi menggunakan teknik bounding box serta proses augmentasi data menggunakan platform Roboflow sehingga jumlah dataset meningkat menjadi 856 citra. Proses pelatihan model dilakukan menggunakan algoritma YOLOv5 untuk mendeteksi objek tandan kelapa sawit pada gambar dan menghasilkan nilai precision sebesar 0.859, recall sebesar 0.673, mAP@0.5 sebesar 0.767, serta mAP@0.5:0.95 sebesar 0.542. Model yang telah dilatih kemudian dikonversi ke dalam format TensorFlow Lite dan diintegrasikan ke dalam aplikasi mobile berbasis Flutter. Aplikasi ini memungkinkan pekerja mengambil gambar tandan kelapa sawit dan melakukan proses deteksi secara otomatis langsung pada perangkat mobile tanpa bergantung pada server eksternal. Sistem juga memanfaatkan layanan Firebase untuk autentikasi pengguna, penyimpanan data, serta proses verifikasi oleh mandor sehingga diharapkan dapat meningkatkan efisiensi dan transparansi dalam proses pencatatan hasil panen kelapa sawit.

Rincian Artikel

Bagian
Articles

Referensi

[1] A’isyah, S. (2022). BURUH TANI DALAM HUKUM POSITIF INDONESIA:: Sebuah Kajian Perundangan untuk Upaya Gagasan Perlindungan. MAQASHID Jurnal Hukum Islam. https://ejournal.alqolam.ac.id/index.php/maqashid/article/view/792Bereiter-Hahn, J. (1990). Behavior of mitochondria in the living cell. International review of cytology, 122, 1-63.
[2] Alhaji, A. M., Almeida, E. S., Carneiro, C. R., Silva, C. A. S. da, & ... (2024). Palm Oil (Elaeis guineensis): A Journey through Sustainability, Processing, and Utilization. Foods, Query date: 2024-12-31 22:04:02. https://www.mdpi.com/2304-8158/13/17/2814.
[3] Alsuwaylimi, A. A., Alanazi, R., Alanazi, S. M., & ... (2024). Improved and Efficient Object Detection Algorithm based on YOLOv5. … , Technology &Applied …. https://etasr.com/index.php/ETASR/article/view/7386.
[4] Ghofar, A., Rosjidi, M., Setiadi, S., Iswantini, D., & ... (2024a). Study on Palm Oil Cultivation and Processing Technology to Support The Biofuel Program (Optimization for The Development of Palm Oil-Based Energy Plantation in …. … Series: Earth and …. https://doi.org/10.1088/1755-1315/1364/1/012011.
[5] Ghofar, A., Rosjidi, M., Setiadi, S., Iswantini, D., & ... (2024b). Study on Palm Oil Cultivation and Processing Technology to Support The Biofuel Program (Optimization for The Development of Palm Oil-Based Energy Plantation in …. … Series: Earth and …. https://doi.org/10.1088/1755-1315/1364/1/012011
[6] Gustrinita, S., Agustin, S., Firmansyah, Y., & ... (2024). TFOOD (TensorFlow Object Detection) Application Design Smart Solution to Create Digital Communication Tools for People with Deaf and Speech Disabilities. … Humanities and Social …, Query date: 2024-12-31 21:47:42. https://www.aksaqilajurnal.com/index.php/aihss/article/download/541/377.
[7] Harahap, F. L., Andrianto, R., & ... (2023). Pengembangan Aplikasi Kalkulator Multifungsi Menggunakan Flutter Framework. Jurnal Penelitian …. http://jurnal.itbsemarang.ac.id/index.php/JPTIS/article/view/815
[8] Harianti, I., & Anggono, B. D. (2024). Legal Transformation: Realizing Nomocracy and Transparency in the Field of Land Distribution in Indonesia. Jurnal Indonesia Sosial …. https://jist.publikasiindonesia.id/index.php/jist/article/view/1086
[9] Huda, M. I. M. (2024). Comparative Analysis of Leadership and Economic Impact of the Palm Oil Industry in Indonesia. Business and Economic Research. https://macrojournal.org/index.php/ber/article/view/217
[10] Jamil, M., & Asrol, M. (2024). Palm Oil Sustainability: Current and Further Potential Research to Adopt Sustainable Development Goals 2030. IOP Conference Series: Earth and …. https://doi.org/10.1088/1755-1315/1324/1/012077
[11] N. Ruseno, F. Ernawan, M. A. Salamat, M. S. Arshad, and H. Amnur, “An Enhanced YOLOv8 Model Integrated with CNN for Real-Time Vehicle License Plate Recognition,” JOIV : International Journal on Informatics Visualization, vol. 10, no. 1, doi: http://dx.doi.org/10.62527/joiv.10.1.5145.
[12] Liu, C., Wang, G., Xu, L., Qu, L., Zhang, H., Tian, L., Li, C., & ... (2024). Design of an Object Recognition Network Based on YOLOv5s for Lightweight Deep Information Extraction and Fusion of Deep and Shallow Layer Information. researchsquare.com. https://www.researchsquare.com/article/rs-4501888/latest
[13] Muflikhah, L., Mahmudy, W. F., & Kurnianingtyas, D. (2023). Machine Learning. books.google.com. https://books.google.com/books?hl=en&lr=&id=tu_uEAAAQBAJ&oi=fnd&pg=PR4&dq=%22lailil+muflikhah%22+machine+learning&ots=_RrBRTwgLt&sig=2hkZQVPhcSJIy6sC5cX2wPxzRXU
[14] Nuwara, Y., Wong, W. K., & Juwono, F. H. (2022). Modern Computer Vision for Oil Palm Tree Health Surveillance using YOLOv5. https://ieeexplore.ieee.org/abstract/document/10010668/
[15] Nuwara, Y., Wong, W. K., & Juwono, F. H. (2022). Modern Computer Vision for Oil Palm Tree Health Surveillance using YOLOv5. . https://ieeexplore.ieee.org/abstract/document/10010668/