Analisis Multivariat Performa Principal Polar Spectral Indices Pada Eo-1 Hyperion : Studi Kasus Pemetaan Lahan Padi

Suryoprayogo, Hertanto (2017) Analisis Multivariat Performa Principal Polar Spectral Indices Pada Eo-1 Hyperion : Studi Kasus Pemetaan Lahan Padi. Magister thesis, Universitas Brawijaya.

Abstract

EO-1 Hyperion Hyperspectral Remote Sensing banyak digunakan dalam monitoring vegetasi karena memiliki resolusi spectral yang tinggi. Namun demikian, beberapa permasalahan umumnya muncul dalam mengolah data Hyperspectral. Dalam penelitian ini algoritma Pixel Purity Index digunakan dalam mengatasi permasalahan tersebut dalam ekstraksi endmember, spectral unmixing serta mengurangi dimensionalitas pada data. Dalam pemetaan vegetasi, NDVI dan Tasseled Cap merupakan metode yang paling banyak diunakan. Namun terdapat limitasi pada kedua metode tersebut dimana keduanya hanya menggunakan stand-alone model yang hanya berelasi pada komposisi satu dimensi. Principal Polar Spectral (PPS) merupakan high-order index set yang dikembangkan pada Landsat 7 ETM+ yang mampu mengobservasi level kandungan pada vegetasi dan soil. Dalam penelitian ini, PPS dikembangkan pada EO-1 Hyperion dalam meneliti sensitivitas PPSG terhadap endmember padi dengan membagi data EO-1 Hyperion ke dalam tiga model pendekatan berdasarkan korelasi Landsat 7 ETM+ wavebands. Hasil klasifikasi menggunakan SVM menunjukkan keseluruhan akurasi yang rendah sebesar 9.76%, dan Kappa Coefficient = 0.0636. Penelitian lebih lanjut diperlukan dalam meneliti PPS indeks dengan objek vegetasi lain dan serta perlu dilakukan penelitian mengenai keunggulan PPS dalam menangkap variasi spectral pada berbagai objek yang lebih luas menggunakan metode VI lain.

English Abstract

EO-1 Hyperion Hyperspectral Remote Sensing is widely used in vegetation monitoring because its high spectral resolution. However, some problems generally arise in Hyperspectral data processing. In this study, the Pixel Purity Index algorithm is used to overcome the problem in endmember extraction, spectral unmixing as well as reduce the data dimensionality. In vegetation mapping, NDVI and Tasseled Cap are the most widely used methods. There are, however, Limitations on both methods which use stand-alone models and relate only to one-dimensional compositions. Principal Polar Spectral (PPS) is a highorder index set developed on Landsat 7 ETM + that is able to observe the level of content in vegetation and soil. In this study, PPS was developed at EO-1 Hyperion in investigating PPSG sensitivity to paddy endmember, by divided three model EO-1 Hyperion data approach based on Landsat 7 ETM+ wavebands correlation. The result of classification using SVM shows a low overall accuracy of 9.76%, and Kappa Coefficient = 0.0636. Further research is needed in examining the PPS index with other vegetation objects and as well as research on the superiority of PPS in capturing the spectral variation in wider object by using another VI method.

Item Type: Thesis (Magister)
Identification Number: TES/005.43/SUR/a/2017/041707622
Uncontrolled Keywords: REMOTE SENSING - COMPUTER PROGRAM, MULTIVARIATE ANALYSIS, SPECTRAL LINE BROADING, HYPERIION (computer)
Subjects: 600 Technology (Applied sciences) > 621 Applied physics > 621.3 Electrical, magnetic, optical, communications, computer engineering; electronics, lighting > 621.36 Optical engineering > 621.367 Technological photography and photo-optics > 621.367 8 Remote sensing technology
Divisions: S2/S3 > Magister Ilmu Komputer, Fakultas Ilmu Komputer
Depositing User: Nur Cholis
Date Deposited: 29 Aug 2017 04:51
Last Modified: 30 Dec 2020 13:26
URI: http://repository.ub.ac.id/id/eprint/1840
[thumbnail of bagian depan.pdf] Text
bagian depan.pdf
Restricted to Repository staff only

Download (226kB)
[thumbnail of BAB 1.pdf] Text
BAB 1.pdf
Restricted to Repository staff only

Download (96kB)
[thumbnail of BAB 2.pdf] Text
BAB 2.pdf
Restricted to Repository staff only

Download (1MB)
[thumbnail of BAB 3.pdf] Text
BAB 3.pdf
Restricted to Repository staff only

Download (726kB)
[thumbnail of BAB 4.pdf] Text
BAB 4.pdf
Restricted to Repository staff only

Download (2MB)
[thumbnail of BAB 5.pdf] Text
BAB 5.pdf
Restricted to Repository staff only

Download (3MB)
[thumbnail of BAB 6.pdf] Text
BAB 6.pdf
Restricted to Repository staff only

Download (838kB)
[thumbnail of BAB 7.pdf] Text
BAB 7.pdf
Restricted to Repository staff only

Download (62kB)
[thumbnail of Daftar Pustaka.pdf] Text
Daftar Pustaka.pdf
Restricted to Repository staff only

Download (72kB)

Actions (login required)

View Item View Item