<![CDATA[Diseases in tomato plants, such as mosaic virus and yellow leaf curl virus, can significantly reduce crop yields. Therefore, early detection based on artificial intelligence (AI) presents a strategic solution to improve the efficiency of plant disease identification. This study aims to develop and evaluate a classification model using Support Vector Machine (SVM) for the automatic and accurate detection of tomato leaf diseases. SVM is selected as the primary classification method due to its ability to handle high-dimensional data with better computational efficiency compared to Convolutional Neural Network (CNN) and Random Forest. The dataset used is the PlantVillage Tomato Leaf Dataset from Kaggle, consisting of 600 images categorized into three classes: healthy tomato leaves, leaves affected by mosaic virus, and leaves affected by yellow leaf curl virus. The research stages include data preprocessing such as image normalization, dataset splitting (80% training, 20% testing), and undersampling to address class imbalance. The SVM model is trained using various kernels and evaluated using accuracy, precision, recall, and F1-score metrics. The results show that the SVM model achieves an accuracy of 98.33%, demonstrating its effectiveness in detecting tomato plant diseases. Therefore, this model can be implemented in smart agriculture systems to enhance early disease detection and assist farmers in optimizing crop yields.]]>
Copyrights © 2025
