Guava Leaf Disease Classification Using Support Vector Machine

Abstract

Agribusiness is the main energy to develop nourishments, raising a human's life and creatures by delivering wanted plant items. In harvesting of things like rice or guava, India is one among the biggest players. Recognizable proof of the plant diseases is the way to forestalling the misfortunes in the yield and amount of the item. Wellbeing checking and disease discovery on plant is extremely basic for supportable horticulture. It is hard to screen the plant maladies physically. It requires enormous measure of work, ability in the plant diseases, and furthermore require the over the top preparing time. Thus, image handling is utilized for the discovery of plant diseases. This work proposes a philosophy for identifying guava leaf maladies early and precisely utilizing image preparing methods and Support Vector Machine (SVM). The proposed framework comprises of following stages like Image pre-preparing, Image segmentation, cluster of an image utilizing k-means, extraction using Gray Level Co-Occurrence Matrix (GLCM). Then the classification of the image is carried out with SVM classifier. In contrasted with existing framework, the proposed framework essentially recognizes the plant leaf disease at an early sickness and improve the accuracy to 98.17%.