Disease Detection in Plants using Convolutional Neural Network
Daida Sharath, Sushant Kumar, M G Rohan, Akhilesh Akhilesh, Kunchapu suresh, C Prathap
Abstract
Majority population of the world depends on agriculture as their primary occupation to earn their income. If any problems occur in that primary sector, it is going to affect the livelihood of the population adversely. Hence, it is necessary to maintain the proper balance in the agriculture sector by preventing the same from adverse effects like drought, plant diseases etc. In agriculture sector, especially horticulture yield more income to the farmers than other crops. These crops are prone to many diseases very easily and manual detection of disease in crops is very much difficult in the early stage. To avoid errors due to manual detection of diseases, Machine learning methods are used. Image processing is done by capturing the infected region of image. The infected image is provided for enhancement followed by image segmentation. then, the segmented image is given as input for the classification using convolutional neural network. This entire process is going to take place over an Android platform. This Android application helps the farmers to identify the disease easily. This paper proposes the techniques involved in the detection of diseases in the very early stage using image processing and the classification is made using convolutional neural network. After the identification a proper solution are going to provide to combat the infection in a very early stage to the farmers. This will increase the income of the farmers.