Tomato plant disease classification in digital images using classification tree

作者：Hiteshwari Sabrol, Kulkarni Satish · 年份：2016 · DOI：10.1109/iccsp.2016.7754351 · 被引用次数：207 · 研究领域：Smart Agriculture and AI、Leaf Properties and Growth Measurement、Spectroscopy and Chemometric Analyses

The applications based on image processing for plant disease recognition and classification is the wide area of research these days. These applications are useful for timely recognition of plant disease. The disease like fungal, bacterial and virus are the destructive disease for any plant. In the study, five types of tomato diseases i.e. tomato late blight, Septoria spot, bacterial spot, bacterial canker, tomato leaf curl and healthy tomato plant leaf and stem images are classified. The classification conducted by extracting color, shape and texture features from healthy and unhealthy tomato plant image. The feature extraction process is done after the segmentation process. Extracted features from segmented images fed to classification tree. Finally, the disease classification was based on these six different types of classes. The classification of six types of tomato images yielded overall 97.3% of classification accuracy.