Please use this identifier to cite or link to this item: http://repository.ipb.ac.id/handle/123456789/71961
Title: Analisis Serangan Virus Gemini pada Cabai Merah (Capsicum annum L.) Berbasis Visual dengan Segmentasi Bayes
Authors: Solahudin, Mohamad
Manaf, Romi
Issue Date: 2014
Publisher: Bogor Agricultural University (IPB)
Abstract: Kebutuhan cabai semakin meningkat dengan peningkatan jumlah penduduk. Salah satu kendala rendahnya produktivitas cabai disebabkan oleh infeksi virus yang disebut virus Gemini. Tanaman yang terinfeksi menunjukkan gejala berupa klorosis pada daun, daun keriting dan menguning. Infeksi virus berpotensi menurunkan hasil sebesar 20 – 100%. Melihat kondisi tersebut, maka diperlukan metode pemantauan dan evaluasi serangan virus yang baik untuk menentukan tindakan yang harus dilakukan. Tujuan dari penelitian ini adalah menganalisis lahan tanaman cabai merah yang terkena serangan virus. Pengambilan citra dilakukan melalui foto udara menggunakan multicopter, citra hasil pemotretan selanjutnya di analisis dengan program pengolah citra untuk mengevaluasi tingkat serangan virus Gemini yang terjadi di lahan. Metode segmentasi Bayes digunakan untuk menentukan tingkat serangan virus pada tanaman individual dengan masukan 3 dimensi berupa komponen warna (R, G, B) dan 4 target segmentasi yaitu tanaman sehat, terserang virus ringan, terserang virus sedang, dan terserang virus akut. Hasil training menggunakan 40 sampel, diperoleh nilai akurasi 100%. Uji coba validasi tanaman individual dengan sampel 240 tanaman untuk 4 kelas diperoleh nilai akurasi 64.24%, validasi dengan target 2 kelas 91.25%, validasi tingkat serangan dengan satuan lahan diperoleh ketelitian 95.58%.
Chili demand in Indonesia has increased in line with increases of population. One of the constraints of low productivity of chili caused by a viral infection called Gemini virus. Infected plants showed symptoms of chlorosis on leaves, leaf curling and yellowing. Viral infections potentially reduce yields by 20-100%. Due to these conditions, then necessary good methods of monitoring and evaluation virus attack to determine what actions to take. The purpose of this study is to analyze the chili plants field were exposed by virus. Images taken through aerial photography via multicopter, the image captured in the subsequent analysis with image processing program to evaluate the level of Gemini virus attacks that occurred on field. Bayesian segmentation method is used to determine the level of virus attacks on individual plants with a 3-dimensional input color components (R, G, B) and 4 segmentation target i.e. healthy plants, mild virus attacks, middle virus attacks, and acute virus attacks. Training results using 40 samples, obtained 100% accuracy rate. Validation test of individual plants with 240 plant samples for 4 classes obtained 64.24% accuracy rate, validation test of individual plants for 2 target class of attack obtained 91.25%, validation test of total fields attack obtained 95.58% accuracy.
URI: http://repository.ipb.ac.id/handle/123456789/71961
Appears in Collections:UT - Agricultural and Biosystem Engineering

Files in This Item:
File Description SizeFormat 
F14rma.pdf
  Restricted Access
Full Text1.88 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.