Please use this identifier to cite or link to this item:
https://repositori.uma.ac.id/handle/123456789/24590
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Muliono, Rizki | - |
dc.contributor.author | Pratama, Bambang | - |
dc.date.accessioned | 2024-07-11T04:05:50Z | - |
dc.date.available | 2024-07-11T04:05:50Z | - |
dc.date.issued | 2024-03-28 | - |
dc.identifier.uri | https://repositori.uma.ac.id/handle/123456789/24590 | - |
dc.description | 95 Halaman | en_US |
dc.description.abstract | Indonesia merupakan negara penghasil utama minyak kelapa sawit yang memproduksi lebih dari 44%. Kelapa sawit sendiri merupakan sumber utama minyak nabati yang memberikan kontribusibesar minyak nabati dalam perdagangan dunia. Dalam aktivitas budidaya kelapa sawit terdapat penurunan hasil produksi yang dialami para petani maupun perusahaan yang disebabkan oleh salah satunya yaitu hama pada tanaman kelapa sawit.Terdapat banyak hama pada tanaman kelapa sawit salah satunya hama UPDKS ( Ulat Pemakan Dan Kelapa Sawit) seperti ulat api setora nitens, ulat kantong metisa plana dan ulat api setothosea asigna. Identifikasi dan klasifikasi hama tanaman kelapa sawit secara manual oleh para petani maupun perusahaan membutuhkan waktu yang cukup tinggi. Oleh karena itu perlu dilakukan analisis kinerja arsitektur GoogleNet dalam konteks klasifikasi hama pada tanaman kelapa sawit. Analisis ini diharapkan dapat memberikan pemahaman lebih dalam tentang keunggulan dan keterbatasan dari kinerja arsitektur GoogleNet dalam mengenali dan mengklasifikasikan hama pada tanaman kelapa sawit. Penerapan metode Convolutional Neural Network dengan arsitektur GoogleNet menghasilkan peforma yang baik dengan jumlah dataset sebanyak 3076 gambar, yang terdari dalam 3 class. Pada skenario model yang menggunakan hyperparameter dengan jumlah epoch 25, batchsize 32, optimizer RMSprop, learning rate 0.001didapatkan hasil akurasi sebesar 100% dengan nilai persisi (Precision) sebesar 100%, recall100% dan f1.score 100%. Indonesia is a major palm oil producing country producing more than 44%. Palm oil itself is the main source vegetable oil which makes a large contribution to vegetable oil in world trade. There has been a decline in oil palm cultivation activities production results experienced by farmers and companies caused by one of which is pests on oil palm plants. There are many pests on One of the pests of oil palm plants is UPDKS (Coconut Eating Caterpillars). Palm oil) such as setora nitens fire caterpillars, metisa plana bagworms and fire caterpillars setothosea asigna. Identification and classification of oil palm plant pests manual work by farmers and companies requires sufficient time tall. Therefore, it is necessary to analyze the performance of the GoogleNet architecture in depth context of pest classification in oil palm plants. It is hoped that this analysis can provide a deeper understanding of the advantages and limitations of GoogleNet architecture performance in recognizing and classifying pests oil palm plants. Application of the Convolutional Neural Network method with GoogleNet architecture produces good performance with a large number of datasets a total of 3076 images, which are contained in 3 classes. In the model scenario using hyperparameters with number of epochs 25, batchsize 32, optimizer RMSprop, learning rate 0.001, obtained accuracy results of 100% with value precision (Precision) of 100%, recall100% and f1.score 100%. | en_US |
dc.language.iso | id | en_US |
dc.publisher | Universitas Medan Area | en_US |
dc.relation.ispartofseries | NPM;198160039 | - |
dc.subject | Klasifikasi | en_US |
dc.subject | kelapa sawit | en_US |
dc.subject | deep learning | en_US |
dc.subject | dan googlenet | en_US |
dc.subject | and googlenet | en_US |
dc.title | Klasifikasi Hama Serangga pada Perkebunan Kelapa Sawit | en_US |
dc.title.alternative | Classification of Insect Pests in Oil Palm Plantations | en_US |
dc.type | Thesis | en_US |
Appears in Collections: | SP - Informatic Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
198160039 - Bambang Pratama - Chapter IV.pdf Restricted Access | Chapter IV | 745.74 kB | Adobe PDF | View/Open Request a copy |
198160039 - Bambang Pratama - Fulltext.pdf | Cover, Abstract, Chapter I, II, III, V, Bibliography | 2.51 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.