Precision Agriculture: Federated Learning CNNs Aloe Vera Leaf Disease Analysis

In order to identify and classify illnesses of aloe vera leaves more effectively, this research is the first to use Federated Learning (FL) in conjunction with Convolutional Neural Networks (CNN). It encompasses a holistic strategy, tackling four tiers of leaf diseases using datasets from six separate customers. The novel approach makes use of Federated Learning's collaborative power to combine different client ideas, making the model more resilient and flexible while protecting client data. To skillfully use a convolutional neural network (CNN) architecture to analyze the datasets and identify patterns that correspond to the four stages of illness severity: mild (1-25%), moderate (26–50%), advanced (51–75%), and severe (76–100%). Metrics like accuracy, precision, recall, F1-score, and support were used to assess the model's performance. The results showed that Client yp_6 was very good at illness categorization; she achieved an F1-score of 96.42 and recall and precision of an incredible 96.49% and 96.37%, respectively. The support value of 1462.50 and the excellent accuracy of 0.98 provided additional proof of its effectiveness. Clients yp_4 and yp_5 also showed impressive skills; they averaged 88.50 on the macro and 88.63 on the micro levels, respectively. Metrics like these demonstrated how well they dealt with diverse data and how well they balanced performance across different severity classifications.On the other hand, clients yp_1 to yp_3, while competent, pointed out areas that might be improved upon with somewhat lower averages, which allowed us to spot chances for refining the model. This study proves without a reasonable doubt that convolutional neural networks (CNNs) and a Federated Learning strategy work wonders for the fast, accurate, and privacy-preserving diagnosis of illnesses affecting aloe vera leaves. This innovative application tackles a pressing problem in agriculture and provides a collaborative, scalable solution that might completely change the way plant diseases are detected.