Artificial Intelligence in Smart Agriculture: Modified Evolutionary Optimization Approach for Plant Disease Identification

Abstract

Apple leaf disease is a critical factor affecting the production and consistency of apples. Typically, the current diagnostic equipment requires significant time to diagnose diseases; thus, farmers also overlook the best opportunity to avoid and cure diseases. Detecting apple leaf diseases is a significant research issue, and its primary goal is to find an appropriate technique for diagnosing leaf diseases. This article attempted to suggest a way to diagnose apple plant leaf disease using the Deep Neural Network (DNN). The architecture of the PDDS (Plant Disease Detection System) is planned. The Robust Speed Up Feature (SURF), which allows achieving greater identification and classification precision, is used to remove functionalities and to refine the Modified Grasshopper Optimization Algorithm (MGOA). Classification parameters such as accuracy, retention, F-measure, mistake, and accuracy are measured, and a comparative review shows the efficiency of the proposed approach.