SISTEM KLASIFIKASI CITRA DAUN UNTUK IDENTIFIKASI PENYAKIT TANAMAN SEMANGKA BERBASIS KECERDASAN BUATAN

Abstract

As a widely cultivated horticultural crop, watermelon plays an important role in supporting the national economy due to its high demand in international markets. However, leaf diseases that are not promptly identified often lead to reduced crop yields. To solve this problem, technology plays a crucial role in supporting the agricultural industry. The purpose of this study is to design a system for identifying watermelon leaf diseases using the Convolutional Neural Network (CNN) method. A literature review was conducted to explore relevant methods in plant disease identification. Images of watermelon leaves were collected and processed through several stages, including preprocessing, CNN model training, and performance evaluation. The model was trained on a three-class dataset with proportions of 80 percent for training, 10 percent for validation, and 10 percent for testing. Evaluation was performed using metrics such as accuracy, precision, recall, and confusion matrix to assess the model’s classification performance. The training results showed a gradual improvement in performance over 50 epochs, achieving a final training accuracy of 98.75 percent and a validation accuracy of 97.98 percent. This system demonstrates good performance in recognizing new images and has the potential to serve as an effective tool for identifying diseases in watermelon leaves.

Keywords: Artificial Intelligence, Convolutional Neural Network, Plant Disease Classification, Watermelon Plant