Identifying and Classifying Disease in Cassava Leaves with Resnet

Manihot esculenta is an essential staple crop in numerous developing nations. However, its yield is significantly hindered by various leaf diseases, such as Cassava Bacterial Blight (CBB), Cassava Brown Streak Disease (CBSD), Cassava Green Mottle (CGM), and Cassava Mosaic Disease (CMD). [1] Traditional methods of disease detection are time-consuming and require expert knowledge, making them impractical for many smallholder farmers.[3] This paper examines the application of Convolutional Neural Networks (CNNs) for the automated identification and classification of diseases in cassava leaves. It utilizes a pre-trained ResNet50 architecture, known for its deep layers and residual learning capabilities, to improve accuracy in identifying disease symptoms from leaf images [5]. The dataset comprises images of healthy and diseased cassava leaves, which are preprocessed through normalization and augmentation techniques to enhance model performance. The ResNet50 model is fine-tuned using learning to adapt to our specific classification task. The approach presented in this paper achieves high accuracy, highlighting its promise as an efficient tool for early and precise diagnosis of cassava diseases. This system can significantly aid farmers in timely disease management, ultimately enhancing cassava crop yields and food security in affected regions.