An Efficacious Detecting Tomato Leaf Disease Using RCOA-Based RNN Method

Plant diseases significantly affect both the world's agricultural economy and food security on a worldwide scale. The likelihood of establishing efficient control measures is increased by their early discovery and classification. Convolutional Neural Networks (CNN)-based categorization of tomato leaf diseases using RGB photos has shown encouraging results in several recent research. Despite their usefulness, CNN models have limitations. Sometimes, they fail to focus on the specific areas affected by plant disease and instead may include irrelevant backgrounds or healthy plant parts in their categorization. This research introduces a new approach for identifying diseased areas and extracting relevant features for illness classification using an Rider Chicken Optimization Algorithm (RCOA)-based Recurrent Neural Network (RNN). Compared to traditional CNN techniques, the RNN-based approach is more robust and can better generalize to unknown crop species that are affected. The classification is based solely on distinct features to achieve the highest accuracy. The proposed RCOA-trained RNN classifier is used to classify diseases. To improve the accuracy of the classification results, fictional computing combines the premise of rider optimization with the hierarchical and swarming conduct of chickens to manage huge volumes of data. The suggested RCOA-based RNN is capable of precisely finding infectious illnesses by analyzing the area of focus with 98.8% accuracy in tomato leaves.