KLASIFIKASI PENYAKIT DAUN PADA TANAMAN SINGKONG MENGGUNAKAN METODE ARTIFICIAL NEURAL NETWORK (ANN)

Hutabarat, Alfredo and Haris, Abdul and Praptini, Puji Catur Siswi (2024) KLASIFIKASI PENYAKIT DAUN PADA TANAMAN SINGKONG MENGGUNAKAN METODE ARTIFICIAL NEURAL NETWORK (ANN). Diploma thesis, ITPLN.

[thumbnail of 202031039_Alfredo Hutabarat_Revisi_Skripsi_ALFREDO Hutabarat.pdf] Text
202031039_Alfredo Hutabarat_Revisi_Skripsi_ALFREDO Hutabarat.pdf
Restricted to Repository staff only

Download (3MB)

Abstract

Daun singkong adalah bagian dari tanaman singkong yang sering dimanfaatkan sebagai bahan pangan dan pakan ternak. Selain memiliki manfaat ekonomi, daun singkong juga rentan terhadap berbagai jenis penyakit yang dapat menurunkan produktivitas tanaman. Penelitian ini bertujuan untuk mengembangkan sistem klasifikasi penyakit daun singkong menggunakan metode Artificial Neural Network (ANN) dengan arsitektur Learning Vector Quantization (LVQ). Sistem ini dirancang untuk mengenali tiga jenis kondisi daun singkong, yaitu sehat, terinfeksi Bercak Daun Coklat, dan Mosaik Kingkong. Pada pengumpulan data, jumlah data yang digunakan adalah 1004 yang dimana dibagi menjadi data latih sebanyak 799 dan data uji sebanyak 205. Data gambar daun singkong yang digunakan diperoleh dari basis data Kaggle, kemudian diproses melalui tahap pre-processing, pengubahan ukuran gambar, dan klasifikasi menggunakan LVQ. Hasil evaluasi model algoritma Learning Vector Quantization menunjukkan bahwa sistem memiliki akurasi sebesar 80% dengan nilai Mean Squared Error (MSE) sebesar 0,2328431, Root Mean Squared Error (RMSE) sebesar 0,0152591, dan Mean Absolute Error (MAE) sebesar 0,03186275. Dengan hasil ini, sistem diharapkan dapat membantu petani dalam mengklasifikasi penyakit daun singkong secara cepat dan akurat, sehingga dapat meningkatkan efisiensi dalam pengelolaan tanaman singkong.

Cassava leaves are a part of the cassava plant that is often used as food and animal feed. In addition to its economic benefits, cassava leaves are also vulnerable to various types of diseases that can reduce plant productivity. This research aims to develop a cassava leaf disease classification system using the Artificial Neural Network (ANN) method with the Learning Vector Quantization (LVQ) architecture. This system is designed to recognize three conditions of cassava leaves: healthy, infected with Brown Leaf Spot, and Cassava Mosaic Disease. During data collection, the total number of data used was 1004, which was divided into 799 training data and 205 test data. The cassava leaf image data used was obtained from the Kaggle database, then processed through the stages of pre-processing, image resizing, and classification using LVQ. The evaluation results of the Learning Vector Quantization algorithm model show that the system has an accuracy of 80% with a Mean Squared Error (MSE) of 0.2328431, Root Mean Squared Error (RMSE) of 0.0152591, and Mean Absolute Error (MAE) of 0.03186275. With these results, the system is expected to assist farmers in classifying cassava leaf diseases quickly and accurately, thereby increasing efficiency in cassava plant management.

Item Type: Thesis (Diploma)
Uncontrolled Keywords: Daun singkong, klasifikasi penyakit, Learning Vector Quantization, Artificial Neural Network Cassava leaves, disease classification, Learning Vector Quantization, Artificial Neural Network
Subjects: Skripsi
Bidang Keilmuan > Teknik Informatika
Divisions: Fakultas Telematika Energi > S1 Teknik Informatika
Depositing User: Sudarman
Date Deposited: 22 Sep 2025 06:50
Last Modified: 22 Sep 2025 06:56
URI: https://repository.itpln.ac.id/id/eprint/1367

Actions (login required)

View Item
View Item