Potato Leaf Disease Detection Using the Convolutional Neural Network

Abstract

Plant diseases, a momentous challenge which initial phase of identification diseases plays a prominent function in managing the prevalence of infection and improving the farming industry. The study is concerned about an approach for the development of a potatoes leaves disease's recolonization model by utilizing deep learning. the configuration of 80/20 is employed for training the model.The Adam is used as the optimizer, the augmentation techniques like flips, rotations are applied to avoid overfitting problem in order to improve the performance and robustness of the model. Our model obtained significant result with 97% accuracy and this study can be used to accurately assess potato leaf diseases detection. Our proposed model successfully performs classification on three types of potato leaves, including healthy, early blight, and late blight.