Develop Hybrid Wolf Optimization with Faster RCNN to Enhance Plant Disease Detection Performance Analysis

Abstract

Plant diseases appear to have become a major threat to global food security, both in terms of production and supply. In this paper, we present a real-time plant disease that relies on altered deep convolutional neural networks. The plant illness images first were expanded by image processing technologies, resulting in the plant disease sets of data. A Wolf Optimization with Faster Region-based Convolutional Neural Network (WO-FRCNN) system that improved removal characteristics was used to identify plant diseases. The proposed method improved the detection of plant diseases and achieved a precision of 96.32%. Prevention activities achieve the basic rate of 15.01 FPS as the existing methods according to experimental data. This study means that the real detectors Improved WO-FRCNN, which would depend on deep learning. It would be a viable option for diagnosing plant diseases and used for identifying other diseases within plants. The evaluation report indicates that the proposed method provides good reliability.