Show simple item record

dc.contributor.advisorSeminar, Kudang Boro
dc.contributor.advisorSudradjat
dc.contributor.authorChoirudin, Ahmad
dc.date.accessioned2024-05-07T23:47:28Z
dc.date.available2024-05-07T23:47:28Z
dc.date.issued2024
dc.identifier.urihttp://repository.ipb.ac.id/handle/123456789/148955
dc.description.abstractPenelitian dalam mendukung perwujudan pertanian presisi terus dikembangankan. Penelitian ini bertujuan untuk memprediksi nutrisi kalsium, magnesium dan boron pada kelapa sawit di lahan gambut dengan menggunakan analisis citra sentinel 2A. Dataset nutrisi diperoleh dari leaf sampling unit pada perkebunan yang berbeda di lahan gambut Pulau Sumatera dan Kalimantan. Proses analisis meliputi pra-pemrosesan citra, pra-proses data, pelatihan model random forest, validasi dan evaluasi model, serta penyusunan interpretasi nutrisi. Hasil penelitian menunjukkan bahwa model mampu menghasilkan akurasi yang layak pada kalsium dan boron dengan correctness 57.13% dan 61.60%, dan akurasi yang baik pada magnesium sebesar 82.27%. Hasil tersebut diperoleh dari data yang terkonsentrasi pada rentang tingkat kandungan tertentu pada setiap nutrisi. Untuk dapat meningkatkan performa model, penelitian ini perlu dilanjutkan dengan menambahkan data sampel agas dapat mewakili setiap rentang nutrisi, dan mengintegrasikannya dengan variabel lain seperti data iklim, tanah, usia pohon, dan selainnya. Penggunaan citra satelit dengan resolusi yang lebih baik dapat menjadi pilihan yang layak dipertimbangkan.id
dc.description.abstractResearch in supporting the realization of precision agriculture continues to be developed. This study aims to predict calcium, magnesium, and boron nutrients on peatland’s oil palm using sentinel-2A imagery analysis. The nutrient dataset was obtained from leaf sampling units in different plantations on peatlands in Sumatera and Kalimantan. The analysis model include imagery preprocessing, data pre-processing, random forest model training, validation and evaluation model, and nutrient interpretation. The study found that the model achieved a responsible level of accuracy for calcium and boron, with correctness rates of 57.13% and 61.60%, respectively. The model also performed well for magnesium, with an accuracy rate of 82.27%. However, it is important to note that these results were obtained from data that was concentrated within a certain range of nutrient levels. To improve the model's performance, future research should focus on adding sample data that represents a wider range of nutrient levels, as well as integrating other variables such as climate data, soil type, and tree age. The use of satellite imagery with higher resolution may also be worth considering.id
dc.language.isoidid
dc.publisherIPB Universityid
dc.titlePendugaan Kadar Kalsium, Magnesium, dan Boron pada Kelapa Sawit di Lahan Gambut dengan Analisis Citra Satelit Sentinel-2Aid
dc.title.alternativeEstimation of Calcium, Magnesium, and Boron Levels in Oil Palm on Peatlands Using Sentinel-2A Satellite Imagery Analysisid
dc.typeUndergraduate Thesisid
dc.subject.keywordoil palmid
dc.subject.keywordprecision agricultureid
dc.subject.keywordrandom forestid
dc.subject.keywordsentinelid
dc.subject.keywordmachine learningid


Files in this item

Thumbnail
Thumbnail
Thumbnail

This item appears in the following Collection(s)

Show simple item record