A Wide Convolution Network Based Classification Path for Recognition Papaya Leaf Disease

Abstract

The top papaya-producing countries are India, Mexico, Brazil, Indonesia, Nigeria and the Dominican Republic. But in papaya especially leaf curl, foot rot of papaya, papaya ring spot and many other diseases are causing great damage to papaya yield. If the diseases are not determined in the early stages, papaya production will decline. The main goal of this piece of research work is to develop and put into effect an algorithm for diagnosing papaya illnesses at the early stage, which are Papaya leaf curl virus (PaLCuV), Papaya ring spot disease, Anthracnose, Foot rot of papaya and Papaya mosaic disease. This paper presents a technique for papaya disease detection using photograph processing techniques, to understand papaya field diseases from images, primarily based on the color, texture and form of diseased papaya and provide appropriate options to farmers, so that papaya diseases can be prevented at an early stage and as a result higher production of papaya can be achieved. I used the cross-validation technique in this model and get the accuracy of 87% which is more than expectation.