Diagnosa Penyakit Tanaman Tomat pada Citra Daun Menggunakan Metode Convolutional Neural Network (CNN)

Abstrak

A Convolutional Neural Network (CNN) was employed to identify diseases in tomato plants through leaf images. The dataset comprised 10,000 images, divided into three parts: 85% for training, 10% for validation, and 5% for testing. Data preprocessing included resizing and labeling images according to their disease type. The CNN model utilized DenseNet121 for feature extraction, leveraging weights pre-trained on the ImageNet dataset. The testing results showed a validation accuracy of 93%, indicating that the model can accurately identify tomato leaf diseases. This study demonstrates that CNNs can improve the efficiency and effectiveness of plant disease detection compared to traditional methods.