SiPadi: An Android-Based Rice Disease Detection Application Using YOLOv8 with CPMAI Methodology – A Case Study in Kandri Village

As a cornerstone of global food security, rice production faces constant threats from climatic shifts, pests, and rapidly spreading diseases that can devastate entire harvests. This issue is especially severe in rural areas like Kandri Village, Central Java, Indonesia where farmers, using traditional methods and limited resources, struggle with early and accurate plant disease diagnosis, and often resulting in irreversible damage. To address this gap, this study presents an Android-based mobile application for early and real-time rice disease detection. Powered by the YOLOv8 object detection model, the app is the first smartphone tool specifically designed for farmers to instantly identify diseases in their fields. Built using the Cognitive Project Management for AI (CPMAI) framework, the trained YOLOv8 model achieved an F1-score of 87% and a mean Average Precision (mAP@50) of 0.894, validating its effectiveness. This tool empowers farmers through accurate, real-time insights, enabling timely intervention, reducing crop loss, and supporting more sustainable rice cultivation.