Image processing based system for the detection, identification and treatment of tomato leaf diseases

• Published in: Multimedia tools and applications Vol. 82; no. 6; pp. 9431 - 9445
• Main Authors: Rahman, Sami Ur, Alam, Fakhre, Ahmad, Niaz, Arshad, Shakil
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.03.2023; Springer Nature B.V
• Subjects: Algorithms; Annotations; Computer Communication Networks; Computer Science; Crop diseases; Data Structures and Information Theory; Health services; Image processing; Image processing systems; Medical imaging; Multimedia Information Systems; Plant diseases; Special Purpose and Application-Based Systems; Support vector machines; Tomatoes; Tomato crop; Septoria leaf spot; Leaf diseases; Image processing; Early blight, late blight
• ISSN: 1380-7501; 1573-7721
• DOI: 10.1007/s11042-022-13715-0

Disease detection and treatment in tomato plant at its early stage contributes towards better production. It is a natural phenomenon that normally tomato plant got disease and if a proper care and remedial action have not taken on time, it badly affects the respective product quality, quantity or productivity. Health monitoring and disease detection of leaves in tomato crop is very critical and if left untreated, can cause serious problems with the plant and fruit, resulting in large losses, especially in fresh markets. Traditional manual methods of disease detection and treatment is based on naked eye observation which cannot provide accurate and on time information at a very early stage of its attack. This paper presents an image processing based techniques for the automatic detection and treatment of leaf diseases in tomato crop. In the proposed method, 13 different statistical features are calculated from tomato leaves using Gray Level Co Occurrence Matrix (GLCM) algorithm. The obtained features are classified into different diseases using Support Vector Machine (SVM). The processed leaf is compared with the stored features on the basis of which disease is recognized. Data are collected from local tomato crop fields and the dataset is divided into a training set and a test set used in the experiments. Experimental results show that the proposed method provides excellent annotation with accuracy of 100% for healthy leaf, 95% for early blight, 90% for septoria leaf spot and 85% late blight. The proposed method is implemented in the form of a cell phone application.