Leaf Disease Detection Using Deep Neural Network

Abstract

The research has concentrated heavily on the utilization of computer vision tasks to identify and categorise phytopathogens. Despite the fact that because herbs are plants too and are prone to disease, the identification of herb diseases has been limited owing to a large amount of study focusing on plant illnesses. A parsley disease detection identification and classification algorithm has been developed in order to recognise and categorise the parsley leaf spot (PLS) sickness according to severity. With 99.5 percentage accuracy rate in both binaries and inter of the PLS sickness, the suggested method employs a neural network convolutional (CNN)-based computational modeling (DL) model to detect 2000 real-time photographs of parsley leaves mixing healthy and PLS unhealthy images. The proposed model beats state-of-the-art pre-trained models in terms of multi-classifying PLS disease, as shown by comparisons between them and the proposed approach.