Cross-Platform Multimodal Frogeye Leaf Spot Recognition and Classification: Advancing with Cutting-Edge CNN Technologies and Transfer Learning in Diverse Plant Species

Frogeye Leaf Spot, a common agricultural disease, poses a significant threat to crop yields. Using multimodal data from different platforms successfully has been a hurdle for existing research in leaf spot detection. Using cutting-edge convolutional neural network (CNN) and transfer learning techniques, this research presents a Cross-Platform Multimodal Recognition system to solve this issue. With a recall of 0.88, F1-score of 0.90, accuracy of 0.91, and precision of 0.92, the suggested method demonstrates exceptional performance. Our accuracy is 6% higher, F1-score is 8% higher, and recall is 9% higher than the preceding state-of-the-art, according to comparisons with previous studies. Improved Frogeye Leaf Spot detection on many platforms is a direct result of our method's success in combining picture and sensor data. This study improves farmers' ability to identify diseases and provides a solid answer for precision agriculture. Our Cross-Platform Multimodal Recognition system has shown to be effective in enhancing agricultural disease recognition, and the shown advances highlight its potential influence on crop management tactics.