Agricultural Innovation: Leveraging DenseNet-121 for Superior Paddy Disease Management

Affecting rice crops through different pathogens, paddy diseases can produce symptoms including discolouration and slowed development, therefore affecting production and quality. Respected for its deep learning capacity and dense connection that supports effective feature extraction, the DenseNet121 model was refined for paddy disease classification in order to address this problem. This model is quite good at capturing variability in disease pictures and complicated trends. Accuracy, recall, and F1-score analysis helped to evaluate the DenseNet121 model. High accuracy of the model indicated efficient type of disease diagnosis and reduced false positives. Even although memory scores were somewhat lower, signalling room for improvement in catching all positive occurrences, the Fl-score displayed a great balance between accuracy and recall. This reveals the minimum misclassification capacity of the model in several illness class differences. On a more complicated assessment, however, more complete view of misclassifications across various disease types could come from more insights from the confusion matrix. Even more so is the performance of the model improved by customizing it to certain application criteria-such as giving some diseases top priority or minimizing false positives.