Innovative Hybrid Deep Learning Strategy for Detecting and Classifying White Rot in Onions

The disease severity of onion white rot has to be measured carefully and correctly to ensure proper agricultural management of this crop. It is one of the most threatening diseases affecting onions since it is caused by the fungal organism called Sclerotium cepivourum. This imperative calls for the research we introduce, a novel hybrid model combining the ability of Convolutional Neural Networks (CNN) with the explained ability of decision tree (DT). The symbiotic integration tries to enhance the precision of classifying onion white rot intensity to a fine-tuned automated diagnosis. Our study is based on a custom database with 3500 detailed pictures of 6 intensity grades of onion white rot. This heterogeneous database provided inputs to our hybrid model which was fine-tuned to achieve an impressive overall accuracy of 94.82%. The performance of the model's robustness is also explained using a multitude of detailed performance measures such as precision, recall, and F1 score. Our hybrid model proves to be superior in comparison with conventional approaches, as evidenced by both high accuracy and increased visibility in making decisions. The ability to discriminate is essential for stakeholders who want to understand the basis for the assigned severities. Our research goes beyond the limits of academic institutions and it has implications for precision agriculture. Our hybrid model automatically provides accurate estimates of disease severity leading to focused intervention, preventing yield loss, and improving resource exploitation. It also aligns with overall objectives in pursuing sustainable and knowledge-based agriculture.