dc.contributor.advisor | Munibah, Khursatul | |
dc.contributor.advisor | Ardiansyah, Muhammad | |
dc.contributor.author | Aina, Arieza Andriani Nur | |
dc.date.accessioned | 2025-01-23T08:04:50Z | |
dc.date.available | 2025-01-23T08:04:50Z | |
dc.date.issued | 2025 | |
dc.identifier.uri | http://repository.ipb.ac.id/handle/123456789/160956 | |
dc.description.abstract | Sektor hortikultura merupakan salah satu subsektor pertanian yang meliputi berbagai jenis tanaman buah-buahan, sayur-sayuran, bunga, dan tanaman hias. Hortikultura memiliki peran dalam mendukung perekonomian nasional. Pengelolaan sumber daya yang berkelanjutan untuk meningkatkan efektivitas produksi hortikultura dapat memanfaatkan teknologi seperti penginderaan jauh. Citra penginderaan jauh bervariasi dari resolusi rendah hingga tinggi, seperti MODIS (1 km); Landsat (30 m) dan Sentinel-2A (20, 10 m); IKONOS (1– 4 m) dan citra Unmanned Aerial Vehicle (< 1 m). Akuisisi citra untuk lahan pertanian hortikultura yang memiliki petakan kecil dapat menggunakan Unmanned Aerial Vehicle (UAV) sebagai alternatif karena memiliki resolusi yang tinggi. Oleh karena itu, tujuan penelitian ini adalah (1) menganalisis pola spektral tanaman hortikultura berdasarkan citra UAV multispektral; (2) menganalisis tingkat ketelitian klasifikasi tanaman hortikultura dengan metode k-Nearest Neighbor (k-NN) dan Minimum Distance Classification (MDC); (3) memetakan tanaman hortikultura dari kedua metode tersebut. Lokasi penelitian berada di Kebun Percobaan Pasir Sarongge, Kabupaten Cianjur menggunakan titik observasi lapang dan citra UAV multispektral pada akuisisi 6 November 2022. Analisis pola spektral sampel dilakukan dengan mengambil titik sampel dari 11 kelas yang meliputi hortikultura dan non hortikultura yang selanjutnya dibuat kurva karakteristik spektral.vKlasifikasi citra menggunakan metode k-NN dan MDC dinilai berdasarkan overall accuracy melalui matriks kesalahan. Hasil penelitian menunjukkan bahwa respon spektral hortikultura pada band hijau dan merah rendah, pada band tepi merah terjadi peningkatan secara signifikan dan terus meningkat di band NIR. Hasil klasifikasi tanaman hortikultura dengan metode k-NN dan MDC menunjukkan urutan luasan yang berbeda. Urutan terluas pada k-NN adalah cabai, wortel, kentang, pisang, kubis, bawang merah, dan tomat. Sementara itu, urutan terluas pada MDC adalah cabai, pisang, wortel, kubis, kentang, bawang merah, dan tomat. Perbedaan algoritma pengklasifikasian menyebabkan variasi dalam urutan luasan hasil klasifikasi. Overall accuracy untuk k-NN dan MDC masing-masing sebesar 89.37% dan 51.48%. | |
dc.description.abstract | The horticulture sector is one of the agricultural subsectors that
includes various types of fruit plants, vegetables, flowers, and ornamental
plants. Horticulture has a role in supporting the national economy.
Sustainable resource management to increase the effectiveness of
horticultural production can utilize technologies such as remote sensing.
Remote sensing images vary from low to high resolution, such as MODIS (1
km); Landsat (30 m) and Sentinel-2A (20, 10 m); IKONOS (1- 4 m) and
Unmanned Aerial Vehicle images (< 1 m). Image acquisition for
horticultural farmland that has small plots can use Unmanned Aerial
Vehicles (UAV) as an alternative because it has a high resolution. Therefore,
the objectives of this study were (1) to analyze the spectral patterns of
horticultural crops based on multispectral UAV imagery; (2) to analyze the
accuracy of horticultural crop classification using the k-Nearest Neighbor
(k-NN) and Minimum Distance Classification (MDC) methods; (3) to
mapping horticultural crops from both methods. The research location was
at Pasir Sarongge Experimental Farm, Cianjur Regency using field
observation points and multispectral UAV images on November 6, 2022
acquisition. Spectral pattern analysis of samples was carried out by taking
sample points from 11 classes covering horticulture and non-horticulture
which were then made spectral characteristic curves. Image classification
using the k-NN and MDC methods was assessed based on the overall
accuracy through the confusion matrix. The results showed that the spectral
response of horticulture in the green and red bands was low, in the red edge
band there was a significant increase and continued to increase in the NIR
band. The classification results of horticultural crops with k-NN and MDC
methods show a different order of area. The largest order in k-NN is chili,
carrot, potato, banana, cabbage, shallot, and tomato. Meanwhile, the largest
order in MDC is chili, banana, carrot, cabbage, potato, shallot, and tomato.
The difference in classification algorithms causes variations in the order of
area of the classification results. Overall accuracy for k-NN and MDC is
89.37% and 51.48%, respectively | |
dc.description.sponsorship | | |
dc.language.iso | id | |
dc.publisher | IPB University | id |
dc.title | Klasfikasi Tanaman Hortikultura Berbasis Citra UAV Multispektral dengan Pendekatan k-Nearest Neighbor dan Minimum Distance | id |
dc.title.alternative | Classification of Horticultural Commodities based on Multispectral UAV Image with k-Nearest Neighbor and Minimum Distance Approach | |
dc.type | Skripsi | |
dc.subject.keyword | Akurasi | id |
dc.subject.keyword | Drone | id |
dc.subject.keyword | Respon Spektral | id |
dc.subject.keyword | Sensor Parrot Sequoia | id |