Potato Crop Disease Classification Using Convolutional Neural Network

Abstract

Potato is one of the most cultivated and in-demand crops after rice and wheat. Potato farming dominates as an occupation in the agriculture domain in more than 125 countries. However, even these crops are, subjected to infections and diseases, mostly categorized into two grades: (i) Early blight and (ii) Late blight. Moreover, these diseases lead to damage the crop and decreases its production. In this paper, we propose a deep learning-based approach to detect the early and late blight diseases in potato by analyzing the visual interpretation of the leaf of several potato crops. The experimental results demonstrate the efficiency of the proposed model even under adverse situations such as variable backgrounds, varying image sizes, spatial differentiation, a high-frequency variation of grades of illumination, and real scene images. In the proposed Convolution Neural Network Architecture (CNN), there are four convolution layers with 32, 16, and 8 filters in each respective layer. The training accuracy of the proposed model is obtained to be 99.47% and testing accuracy is 99.8%.