Fahmi, Muhamad (2026) RANCANG BANGUN APLIKASI DETEKSI PENYAKIT PADA DAUN UBI JALAR MENGGUNAKAN ALGORITMA CONVOLUTIONAL NEURAL NETWORK (CNN). S1 / D3 thesis, Universitas Kuningan.

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

Download (1MB)
[thumbnail of BAB I] Text (BAB I)
BAB I.pdf

Download (291kB)
[thumbnail of BAB II] Text (BAB II)
BAB II.pdf
Restricted to Repository staff only

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

Download (2MB) | Request a copy
[thumbnail of BAB IV] Text (BAB IV)
BAB IV.pdf
Restricted to Repository staff only

Download (2MB) | Request a copy
[thumbnail of BAB V] Text (BAB V)
BAB V.pdf

Download (233kB)
[thumbnail of DAFTAR PUSTAKA] Text (DAFTAR PUSTAKA)
DAFTAR PUSTAKA.pdf

Download (224kB)
[thumbnail of LAMPIRAN] Text (LAMPIRAN)
LAMPIRAN.pdf
Restricted to Repository staff only

Download (5MB) | Request a copy
Official URL: https://rama.uniku.ac.id

Abstract

Produksi ubi jalar di Kabupaten Kuningan mengalami penurunan yang salah satunya dipengaruhi oleh serangan hama dan penyakit pada daun. Proses identifikasi penyakit yang masih dilakukan secara manual menyebabkan keterlambatan penanganan dan berpotensi menurunkan hasil panen. Penelitian ini bertujuan untuk merancang dan membangun aplikasi deteksi penyakit daun ubi jalar berbasis Android menggunakan algoritma Convolutional Neural Network (CNN) dengan arsitektur MobileNetV3-Large. Metode pengembangan sistem yang digunakan adalah Rapid Application Development (RAD). Dataset yang digunakan terdiri dari 10.000 citra daun ubi jalar yang terbagi ke dalam empat kelas, yaitu daun sehat, bercak daun coklat (Cercospora batatae Zimm), daun penyakit hama ulat grayak, dan tungau puru. Model dilatih menggunakan TensorFlow dan diimplementasikan ke perangkat Android melalui TensorFlow Lite. Hasil pelatihan menunjukkan bahwa model mencapai training accuracy sebesar 98,8% dan validation accuracy sebesar 97,3%, dengan nilai loss yang rendah dan stabil. Hasil pengujian menunjukkan bahwa aplikasi mampu melakukan klasifikasi penyakit daun ubi jalar secara cepat serta membantu proses identifikasi penyakit secara lebih cepat dan praktis. Dengan demikian, aplikasi yang dikembangkan dapat menjadi solusi pendukung bagi petani dalam melakukan deteksi dini penyakit daun ubi jalar di lapangan.

Sweet potato production in Kuningan Regency has declined, partly due to pest and disease infestations affecting the leaves. The manual process of disease identification leads to delays in treatment and has the potential to reduce crop yields. This study aims to design and develop an Android-based sweet potato leaf disease detection application using a Convolutional Neural Network (CNN) algorithm with the MobileNetV3-Large architecture. The system development method used is Rapid Application Development (RAD). The dataset consists of 10,000 sweet potato leaf images divided into four classes: healthy leaves, brown leaf spot (Cercospora batatae Zimm), pest-infested leaves, and mealybugs. The model was trained using TensorFlow and implemented on Android devices via TensorFlow Lite. Training results show that the model achieved a training accuracy of 98.8% and a validation accuracy of 97.3%, with low and stable loss values. Testing results indicate that the application is capable of accurately classifying sweet potato leaf diseases and aids in the identification of diseases more quickly and practically. Consequently, the developed application can serve as a supportive solution for farmers in conducting early detection of sweet potato leaf diseases in the field.

Item Type: Thesis (S1 / D3)
Uncontrolled Keywords: Ubi Jalar, Deteksi Penyakit Daun, Android, CNN, MobileNetV3-Large, Leaf Disease Detection, Sweet Potato, Android
Subjects: Q Science > QA Mathematics > QA76 Computer software
Divisions: Fakultas Ilmu Komputer > S1 Teknik Informatika
Depositing User: S.Kom Muhamad Fahmi
Date Deposited: 21 Jul 2026 07:53
Last Modified: 21 Jul 2026 07:53
URI: https://rama.uniku.ac.id/id/eprint/5887

Actions (login required)

View Item
View Item