Potato leaf disease prediction using RMSProp, Adam and SGD optimizers
Shikha Prasher, Leema Nelson, Mukta Jagdish
Abstract
The advances in agricultural equipment and the utilization of artificial intelligence to the treatment of plant diseases is critical to undertake meaningful research for the development of organic agriculture. Numerous diseases, including such late blight and early blight, have a major effect on the quality and yields of potatoes, and subjective analysis of The several leaf diseases require a lot of time and labour. Despite the high level of knowledge required, automated and efficient recognition of these infections during the blossoming phase can help to boost the output of the potato crop. This research provides a sequential model that identifies the significant features from the dataset utilising pre-trained deep learning model with RMSProp, Adam and SGD optimizer. The convolutional neural network (CNN) model is developed using potato leaf diseases dataset obtained from Kaggle respository. The RMSProp, Adam and SGD optimizers achieve accuracy of 97.6%, 96% and 94.3%.