Penggunaan Data Citra Satelit Untuk Mengidentifikasi Varietas Padi Menggunakan Metode Linear Spectral Unmixing (Studi Kasus: Kecamatan Karangjati, Kabupaten Ngawi)

Kusoiry, Moch Rafli (2023) Penggunaan Data Citra Satelit Untuk Mengidentifikasi Varietas Padi Menggunakan Metode Linear Spectral Unmixing (Studi Kasus: Kecamatan Karangjati, Kabupaten Ngawi). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 03311940000063-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 September 2025. Download (8MB) | Request a copy

Abstract

Padi adalah tanaman yang sangat penting bagi negara Indonesia dikarenakan hampir 95% masyarakatnya mengonsumsi beras sebagai bahan pangan pokok. Demi menjaga tingkat kebutuhan konsumsi dan produksi padi tetap seimbang maka perlu dilakukanya kontrol produksi padi berdasarkan varietasnya pada Kabupaten Ngawi khususnya pada Kecamatan Karangjati. Dalam bidang ilmu Geomatika, Penginderaan Jauh menjadi salah satu media yang dapat dimanfaatkan untuk melakukan identifikasi varietas padi. Salah satu inovasi teknologi Penginderaan Jauh adalah pemanfaatan data citra satelit bersensor multispektral. Piksel yang ada pada citra merupakan piksel campuran dari berbagai objek yang direkam, piksel campuran ini dapat mempengaruhi proses klasifikasi, dalam mengatasi hal tersebut maka salah satu upaya yang dapat dilakukan adalah menggunakan metode LSU (Linear Spectral Unmixing), metode LSU sendiri digunakan untuk mengetahui persentase keberadaan suatu objek murni pada setiap piksel citra berdasarkan nilai spektralnya yang disebut endmember, dalam hal ini endmember yang dimaksud adalah varietas padi. Atas semua latar belakang di atas maka dibentuklah penelitian ini yang dilakukan pada masa tanam padi pada masa generatif (70±hst). Hasil yang didapatkan dari penelitian ini yaitu, di Kecamatan Karangjati, varietas yang mendominasi adalah Inpari 32 HDB dengan total luasan 1095 ha, dan sisanya adalah varietas campuran dari Ciherang, Cibogo dan lain lain dengan total luasan 1184 ha. Uji validasi lapangan dilakukan untuk mengetahui seberapa akurat hasil dari penelitian ini dengan mengambil sampel di lapangan sebanyak 62 titik sampel, lalu dilakukan uji validasi data Confusion Matrix dan Analisis Kappa. Berdasarkan hasil Confusion Matrix didapatkan Overall Accuracy sebesar 85.48% dan Analisis Kappa sebesar 70.6%.
=====================================================================================================================================
Rice is a very important crop for Indonesia, as nearly 95% of its population consumes rice as a staple food. In order to maintain a balance between rice consumption and production, it is necessary to control rice production based on its varieties in Ngawi Regency, particularly in Karangjati District. In the field of Geomatics, Remote Sensing is one of the media that can be utilized to identify rice varieties. One of the innovative technologies in Remote Sensing is the utilization of multispectral satellite image data. The pixels in an image are a mixture of various recorded objects, and this mixed pixel can affect the classification process. To address this issue, one of the approaches that can be employed is the Linear Spectral Unmixing (LSU) method. The LSU method itself is used to determine the percentage of the presence of a pure object in each image pixel based on its spectral value, which is called the endmember, and in this case, the intended endmembers are rice varieties. Based on all the aforementioned background, this research was conducted during the generative stage of rice cultivation (70± days after planting). The findings of this study reveal that the dominant variety in Karangjati District is Inpari 32 HDB with a total area of 1095 ha, while the rest are mixed varieties of Ciherang, Cibogo, and others with a total area of 1184. Field validation tests were conducted to assess the accuracy of the research results by collecting 62 sample points in the field. Subsequently, the Confusion Matrix and Kappa Analysis were employed for data validation. Based on the Confusion Matrix results, the Overall Accuracy obtained was 85.48%, and the Kappa Analysis yielded a value of 70.6%

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Endmember, Kecamatan Karangjati, Linear Spectral Unmixing, Penginderaan Jauh, Varietas Padi; Endmember, Karangjati District, Linear Spectral Unmixing, Remote Sensing, Rice Varieties.
Subjects:	G Geography. Anthropology. Recreation > G Geography (General) G Geography. Anthropology. Recreation > GA Mathematical geography. Cartography > GA102.4.R44 Cartography--Remote sensing S Agriculture > S Agriculture (General)
Divisions:	Faculty of Civil, Planning, and Geo Engineering (CIVPLAN) > Geomatics Engineering > 29202-(S1) Undergraduate Thesis
Depositing User:	Moch Rafli Kusoiry
Date Deposited:	03 Aug 2023 03:33
Last Modified:	03 Aug 2023 03:33
URI:	http://repository.its.ac.id/id/eprint/101010

Actions (login required)

View Item