Innovative agricultural diagnosis: DQRR-AFH algorithm model for effective leaf disease prevention and monitoring

In recent decades agricultural decision-making system has played a vital role in the field monitoring process. For these emerging technologies like the Internet of Things (IoT), artificial intelligence (AI), and wireless sensors are utilized for precise data extraction and analysis. However numerous techniques are developed for increasing agricultural production and enhancing operational efficiency. But still, they possess various challenges like lack of accuracy, increased power utilization, and costs. Insects and pathogens cause plant diseases that reduce productivity if not diagnosed at a proper time. Thus this paper develops Deep Q Rapidly-exploring Random tree-based Adaptive Fire hawk (DQRR-AFH) for diagnosing leaf diseases and monitoring them. It comprises various phases including data acquisition; Image processing, segmentation, feature extraction, and classification. Further, the classification is performed by exploring random trees, and the hyperparameters are tuned via the Adaptive Fire Hawk Optimizer algorithm is utilized to enhance the efficiency of the model. The proposed method monitors the soil moisture content and prevents cotton leaf diseases by spraying chemicals on the plants. Multiple cotton leaf images are obtained to verify the performance with various metrics. Compared to conventional methods such as WL-CNN, ECPRC, and DT, the proposed model achieved exceptional performance with an accuracy of 98.88%, precision of 97%, and an F1-score of 99.21%.