Machine Learning-based Disease Classification in Tomato (Solanum lycopersicum) Plants

In Bangladesh, tomato cultivation faces significant challenges due to its susceptibility to various microorganisms, parasites, and bacterial infections. Typically, the early symptoms of these diseases first appear in roots and leaves, complicating timely detection. This study addresses the challenge of timely and accurate detection of diseases in tomato plants, crucial for effective plant protection management. Conventional manual inspection methods are time-consuming and subjective, resulting in delays in implementing necessary protection measures. Therefore, an image processing technique and machine learning algorithms were used for rapid and robust detection of diseases in tomato plant leaves, aiming to streamline the detection process for chemical application responses. A dataset containing 250 images of tomato plant leaves were captured under varying light intensities, eye-level angles, and distances. Image augmentation techniques were applied to increase the dataset, resulting in a total of 529 images. These images were converted to LAB color images and then OTSU algorithm was used to segment leaf images and estimate the percentage of affected diseased areas. Various textural features were also extracted from segmented leaf images to create a training dataset. Machine learning algorithms, including Support Vector Machines (SVM), K-Nearest Neighbors (KNN), and decision trees, were trained and evaluated using this dataset to classify images as healthy or diseased. The Quadratic SVM algorithm provided the highest test accuracy of 97.7% for the dataset. This nondestructive processing holds immense promise for improving disease detection efficiency and reducing losses in tomato production, both locally in Bangladesh and globally.