Artificial intelligence-enabled coconut tree disease detection and classification model for smart agriculture

Abstract

Real-time and accurate plant disease recognition systems help in the development of disease mitigation strategies and ensure food security on a large scale compounded with small-scale economic crop protection. The current research article presents an Artificial Intelligence Enabled Coconut Tree Disease Detection and Classification (AIE-CTDDC) model for smart agriculture. The aim of the presented AIE-CTDDC technique is to classify the coconut tree diseases in a smart farming environment so as to enhance the crop productivity. Firstly, the AIE-CTDDC model applies median filtering-based noise removal technique. Then, the Bayesian fuzzy clustering-based segmentation method is employed for the detection of the affected leaf regions. Besides, the capsule network (CapsNet) method is exploited as a feature extractor. In this study, the Harris Hawks Optimization (HHO) with Gated Recurrent Unit (GRU) model is exploited for the detection of diseases in coconut trees. The experimental analysis was conducted upon AIE-CTDDC model and the outcomes confirmed the better performance of the proposed AIE-CTDDC model over recent state-of-the-art techniques.