Corn Leaves Disease Detection Using Convolutional Neural Networks [CNN

Abstract

A proposed approach for accurately determining prevalent corn diseases involves utilizing a convolutional neural network-based method for corn disease identification. Four different types of maize leaf pictures are produced by adding corrected linear unit activation to the enhanced CNN model, which is used for training and testing functions and the Adam Optimizer, changing the settings, combining processes and cutting down on classifiers. When three different kinds of maize leaf diseases were identified, this model attains an average detection accuracy of 93.75%, which is the maximum precision only attained in the identification of maize illness from leaves with faster rates of training convergence to the highest quality our deliberation. All things considered, this method is quick, easy to use, and offers a dynamic means of identifying leaf maize disease. This will help low-income farmers avoid crop loss and encourage support for the agricultural system across the country. Key Words: Convolutional Neural Network [CNN], Deep Learning, Activation Function, RectifiedLinear Unit [ReLu], SoftMax,