An efficient plant disease prediction model using darner drain fly optimization-based deep convolutional neural network

The primary source of prosperity is agriculture, and it is vital to humankind. However, various plant diseases are the cause of a number of challenges that farmers encounter. One of agriculture's most difficult issues is the early plant disease diagnosis. The overall yield could be pessimistically impacted by diseases if they are not detected in their early stages, which would lower the farmer's earnings. Numerous researchers have developed various state-of-the-art solutions based on various strategies to address this issue. This paper proposed a hybrid model based on a DDFO-based deep Convolutional Neural Network (deep CNN) for automatic plant disease detection. The research's main contribution relies on the darner drain fly optimization (DDFO) that effectively tunes the deep CNN classifier by tuning their parameters using the fitness function. The DDFO optimization is formed by hybridization of dragonfly and mothfly optimizations and the use of the DDFO optimization helps to enhance the performance of the model in plant disease prediction. The feature is efficiently extracted using the Resnet-101 model and the research's superiority is established by assessing the metrics accuracy, sensitivity, and specificity. In comparison to previous methods, the suggested DDFO-based deep CNN classifier was quite effective, achieving 95.41%, 95.37%, and 95.64% when detecting the disease in apples, 95.42%, 95.32%, and 95.71% when detecting the disease in potatoes, and 95.59%, 95.78%, and 95.59% when detecting the disease in strawberries. By significantly improving the accuracy, sensitivity, and specificity of disease identification in various crops, this approach can help farmers mitigate the adverse effects of plant diseases, thereby enhancing crop yield and profitability. These results demonstrate that the developed approach can serve as a robust and efficient tool for early plant disease detection. The proposed model's superior performance underscores its practical applicability and potential to revolutionize agricultural practices through advanced technological integration.