Abstract:
Leaf diseases cause many significant damages and losses to the farmers around the world. Appropriate measures on disease identification should be introduced to prevent the problems and minimize the losses. Technical approaches using machine learning and computer vision are actively researched to achieve intelligence farming by early detection of leaf disease. An analyzer is obviously desirable to aid the farmers in diagnosing what sorts of diseases a leaf has. This dissertation presents the research,design, and implementation of an analyzer which can automatically identify the leaf diseases based on its appearance with some computer vision and machine learning technique. Many experiments and evaluations on different segmentation, feature extractions, and classification methods were done to find the most effective approach. The target group of the user is those who request a free and quick diagnosis of common leaf disease at any time of the day.
