DEVELOPMENT OF A DEEP LEARNING MODEL FOR CASSAVA PLANT DISEASE IDENTIFICATION

Cassava (Manihot esculenta Crantz) is a vital staple crop for millions globally, but its production is severely threatened by Cassava Mosaic Disease (CMD) and Cassava Brown Streak Disease (CBSD), leading to devastating economic and food security losses. To address the challenges of manual disease detection, this study aims to develop a web-based mobile application utilizing Convolutional Neural Networks (CNNs) for rapid and precise disease identification from leaf imagery. A diverse dataset representing healthy and diseased plants was collected, preprocessed, and augmented to optimize model training. The application features a user-friendly interface that allows farmers and agricultural stakeholders to upload images, receive diagnostic results, and access treatment recommendations. Crucially, the app features an offline mode, enabling image capture and initial processing in remote areas with limited internet connectivity. The CNN models were evaluated using standard performance metrics, including accuracy, precision, recall, and F1-score. Ultimately, this digital tool provides an accessible, hardware-independent diagnostic solution that supports early disease management, improves crop yields, and safeguards smallholder farmer livelihoods. Keywords: Cassava, Deep Learning, Convolutional Neural Networks, Plant Disease Identification, Mobile Application, Offline Functionality.