A Novel Approach to Detect and Classify Strawberry Fruit and Leaf Diseases using Deep Learning

Abstract

The research uses robust learning methods to identify powdery mildew and leaf scorch on strawberry fruit and leaves. The suggested method combines ResNet50 and InceptionV3 cognitive neural architectures. This method involves collecting Powdery Mildew and Leaf Scorch datasets and improving them. ResNet50 and InceptionV3 extract distinguishing characteristics and identify powdery mildew and leaf scorch patterns. The combined attributes are sent to a fully connected layer and a softmax classifier to help sort the diseases. An optimised approach modifies the model’s hyperparameters during training for optimal classification performance. Thorough evaluation shows that the combined model outperforms ResNet50, InceptionV3, and other disease identification methods. The merging of features using fusion methods improves representation and disease detection. Research suggests that the reinforcement learning-based approach can detect powdery mildew and leaf scorch in strawberry fruits and leaves with an accuracy rate of 98.00% and a validation rate of 97.34%. This technology has an impact on strawberry farming disease management.