Leveraging Metaheuristic with Deep Learning Models for Enhancing Plant Disease Detection and Classification

Abstract

Agricultural technology is rapidly progressing towards a new paradigm Agriculture 4.0. newlineWith this motivation, artificial intelligence (AI), digitalization, and automation play an integral part newlinein agricultural production, involving pest control and weeding. The application of CV and AI to newlineautomated diagnosis and detection of plant disease is now being comprehensively analyzed since newlinemanual plant disease monitoring is labour-intensive, time-consuming, and tedious. Recently, deep newlinelearning (DL) model is extensively used in recognizing disease. But due to the requirements for newlineclassical neural network model of high quality and quantity of datasets and high hardware resources newlineduring the training process, the training waste so much time that does not facilitate the promotion newlineand use of the model. newlineA real-time and accurate disease detection technique might assist to develop mitigation newlinestrategy to guarantee food security on largescale and economical crop protection on a small scale. newlineMoreover, an accurate disease classification via DL and machine vision provides the basis to achieve newlinethe site-specific application of agrochemicals. At the same time, the introduction of image analysis newlinetool becomes a powerful tool for earlier recognition of plant diseases and continuous monitoring of newlineplant health status. With this motivation, the study presents a set of DL algorithms for the detection newlineand classification of plant leaf disease newline