Deep Learning based Approach for Cassava Leaf Disease Detection

Cassava is a vital staple crop in numerous tropical countries, to enhance plant disease detection using deep learning architecture including ResNet, EfficientNet, Incetion V3 and simple CNN. By levering these advance model significant improve accuracy compared to previous methods. This approach addresses the pressing issue of food insecurity by efficiently identifying diseases in crops. Types of Cassava disease are Cassava Mosaic Disease (CMD), Cassava Green Mottle (CGM), Cassava Brown Streak Disease (CBSD), and Cassava Bacterial Blight (CBB), along with the presence of Whitefly. The proposed work uses Convolution Neural Network (CNN) to identify the leaf diseases to quickly identify the specific disease that is affecting the crop. CNN can automatically detect important features without any human supervision. A vast collection of labelled cassava leaf photos comprising multiple disease classes is utilized for training the CNN model. To advance deep learning architectures, our research integrates techniques for data augmentation and transfer learning to further enchance model performance.