Renawati, Sena (2026) KLASIFIKASI PENYAKIT DAUN TEMBAKAU BERBASIS CITRA DIGITAL MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK. S1 / D3 thesis, Universitas Kuningan.

[thumbnail of ABSTRAK] Text (ABSTRAK)
ABSTRAK.pdf

Download (717kB)
[thumbnail of BAB I] Text (BAB I)
BAB I.pdf

Download (367kB)
[thumbnail of BAB II] Text (BAB II)
BAB II.pdf
Restricted to Repository staff only

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

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

Download (2MB)
[thumbnail of BAB V] Text (BAB V)
BAB V.pdf

Download (340kB)
[thumbnail of DAFTAR PUSAKA] Text (DAFTAR PUSAKA)
DAFTAR PUSAKA.pdf

Download (280kB)
[thumbnail of LAMPIRAN] Text (LAMPIRAN)
LAMPIRAN.pdf
Restricted to Repository staff only

Download (4MB)

Abstract

Tembakau merupakan komoditas pertanian strategis di Indonesia dengan produksi mencapai 238,81 ribu ton pada tahun 2023. Namun, produktivitas tanaman tembakau kerap terganggu oleh berbagai penyakit daun yang sulit diidentifikasi secara manual oleh petani. Penelitian ini bertujuan untuk membangun sistem klasifikasi penyakit daun tembakau berbasis citra digital menggunakan algoritma Convolutional Neural Network (CNN) dengan arsitektur VGG16 melalui pendekatan transfer learning. Dataset yang digunakan terdiri dari 4.000 citra daun tembakau yang terbagi ke dalam empat kelas, yaitu Brown Spot, Leaf Curl, Mosaic, dan Daun Sehat, dengan masing-masing kelas berjumlah 1.000 citra. Dataset tersebut kemudian diproses melalui tahap augmentasi dan pelatihan model dua fase. Model dibangun menggunakan framework TensorFlow dan dikonversi ke dalam format TensorFlow Lite untuk diintegrasikan ke dalam aplikasi berbasis Android. Pengembangan aplikasi dilakukan menggunakan metode Rapid Application Development (RAD) dengan bahasa pemrograman Kotlin. Hasil pengujian menunjukkan bahwa model mampu mengklasifikasikan penyakit daun tembakau dengan akurasi keseluruhan sebesar 98%, dengan nilai Precision, Recall, dan F1-Score rata-rata sebesar 0,98 pada seluruh kelas. Aplikasi yang dihasilkan memungkinkan petani melakukan identifikasi penyakit daun tembakau secara otomatis melalui citra gambar maupun kamera real-time pada perangkat Android.

Tobacco is a strategic agricultural commodity in Indonesia, with production reaching 238.81 thousand tons in 2023. However, the productivity of tobacco plants is often disrupted by various leaf diseases that are difficult to identify manually by farmers. This study aims to develop a digital image-based tobacco leaf disease classification system using the Convolutional Neural Network (CNN) algorithm with the VGG16 architecture through a transfer learning approach. The dataset used consists of 4,000 tobacco leaf images divided into four classes, namely Brown Spot, Leaf Curl, Mosaic, and Healthy Leaf, with each class containing 1,000 images. The dataset was then processed through augmentation and a two-phase model training stage. The model was built using the TensorFlow framework and converted into TensorFlow Lite format to be integrated into an Android-based application. Application development was carried out using the Rapid Application Development (RAD) method with the Kotlin programming language. The test results show that the model is able to classify tobacco leaf diseases with an overall accuracy of 98%, with an average Precision, Recall, and F1-Score of 0.98 across all classes. The resulting application enables farmers to automatically identify tobacco leaf diseases through image uploads or a real-time camera on Android devices, so that it is expected to help farmers detect tobacco leaf diseases quickly, easily, and without requiring special expertise.

Item Type: Thesis (S1 / D3)
Uncontrolled Keywords: Penyakit Daun Tembakau, Transfer Learning, Android, Convolutional Neural Network, VGG16, Tobacco Leaf Disease, Transfer Learning, Android
Subjects: Q Science > QA Mathematics > QA76 Computer software
T Technology > T Technology (General)
Divisions: Fakultas Ilmu Komputer > S1 Teknik Informatika
Depositing User: S.Kom Sena Renawati
Date Deposited: 13 Jul 2026 07:19
Last Modified: 13 Jul 2026 07:19
URI: https://rama.uniku.ac.id/id/eprint/5578

Actions (login required)

View Item
View Item