Abstract
Detection for plant-disease is a crucial problem that needs to be prioritized, for a thriving agriculture and economy. Detecting plant disease using traditional methods are laborious since they call for a lot of effort, time, and knowledge. Recently, academics, researchers, and professionals have all been quite interested in the crucial scientific field of automatic plant disease detection. Presently, Deep Learning (DL) methods, especially those based on Convolutional Neural Network (CNN), These methods have attained widespread applications in plant disease classification. They have either fully solved or partially solved the problems of old-style classification methods and epitomize up-to-the-minute technology in this field. This work includes, train a deep convolutional neural network to detect diseases (or their lack thereof) using a Plant Village dataset from Kaggle consisting of 3152 diseased and well-healthy images of plant leaves acquired under meticulous conditions segmented using proposed segmentation methodology. The trained model performs with an average 96.4% accuracy on a held-out test set, proving that the strategy is feasible.

This publication has 11 references indexed in Scilit: