Please use this identifier to cite or link to this item:
https://repositori.uma.ac.id/handle/123456789/27356
Title: | Klasifikasi Penyakit Tanaman Cabai Menggunakan Googlenet Pada Citra Daun |
Other Titles: | Classification of Chili Plant Diseases Using Googlenet on Leaf Images |
Authors: | Harahap, Jaffar Siddik |
metadata.dc.contributor.advisor: | Sembiring, Arnes |
Keywords: | GoogLeNet;Deep Learning;Image Classification;CNN;Chili Disease;Klasifikasi Gambar;Penyakit Cabai |
Issue Date: | May-2025 |
Publisher: | Universitas Medan Area |
Series/Report no.: | NPM;208160011 |
Abstract: | Penelitian ini bertujuan untuk mengklasifikasikan penyakit pada daun tanaman cabai menggunakan arsitektur GoogLeNet berbasis Deep Learning. Jenis penyakit yang diamati terdiri dari lima kelas, yaitu healthy, leaf curl, leaf spot, whitefly, dan yellowish. Dataset yang digunakan berjumlah 500 gambar yang diambil dari platform Kaggle. Gambar tersebut melewati tahap prapemrosesan seperti resize, augmentasi, dan normalisasi, kemudian dibagi menjadi data latih dan uji. Proses pelatihan dilakukan dengan pendekatan transfer learning menggunakan tiga jenis optimizer, yaitu Adam, SGD, dan RMSprop. Evaluasi kinerja model dilakukan menggunakan matrix accuracy, presicion, recall, F1-score, serta confusion matrix. Hasil penelitian menunjukkan bahwa optimizer Adam menghasilkan accuracy tertinggi sebesar 98.8% dengan nilai validation loss terendah, menunjukkan bahwa metode ini efektif untuk klasifikasi penyakit daun cabai secara otomatis dan akurat. This study aims to classify diseases in chili plant leaves using the GoogLeNet architecture based on Deep Learning. The types of diseases observed consist of five classes, namely healthy, curly leaves, leaf spots, whiteflies, and yellowing. The dataset used consists of 500 images taken from the Kaggle platform. The images go through pre-processing stages such as resizing, augmentation, and normalization, then divided into training and test data. The training process is carried out using a transfer learning approach using three types of optimizers, namely Adam, SGD, and RMSprop. Model performance evaluation is carried out using accuracy, precision, recall, F1-score, and confusion matrices. The results of the study showed that the Adam optimizer produced the highest accuracy of 98.8% with the lowest validation loss value, indicating that this method is effective for automatic and accurate classification of chili leaf diseases |
Description: | 10 Halaman |
URI: | https://repositori.uma.ac.id/handle/123456789/27356 |
Appears in Collections: | SP - Informatic Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
208160011 - Jaffar Siddik Harahap - Fulltext.pdf | Fulltext | 712.79 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.