Transforming Coconut Farming with Deep Learning Disease Detection

Early detection of plant diseases is crucial for effective prevention and control of food insecurity in agriculture. Coconut trees are vulnerable to various pathogens such as fungi, bacteria, viruses, and nematodes, which can cause significant quantitative and qualitative losses. In this study, we suggested a strategy that applies the resulting hybrid model (NNSVCLD), which is created by modifying SVM and CNN, to extract deep features from images and classify out of 5 diseases to which class it belongs. Comprehensive testing is carried out on Kaggle dataset, and performance measures like accuracy, precision, recall, specificity, and F1-rating are assessed and examined. According to the experimental results, the introduced version outperformed the other cutting-edge hybrid learning models in terms of accuracy, precision, recall, and F1 rating for different folds. The proposed method shows accuracy with 98.9% for three, 99.3% for five folds, and 99.4% for 10 folds respectively. The efficacy of the model is further supported by the precision, recall, and F1-scores for each category, which range from 98.7 to 99.4%.