Rice Disease Monitoring
Our ongoing R&D initiative deploys fixed IP-cameras across paddy fields across 4 different farms, streaming live footage via RTSP protocol to Raspberry Pi edge devices for upload to Google Cloud Platform (GCP) through an automated pipeline, enabling scalable, real-time storage and analysis. AI-powered computer vision models then analyze thousands of frames to monitor rice crop growth stages, detect early signs of disease, and track phenological development throughout the season. By combining IoT edge computing with cloud-based machine learning, we empower farmers and researchers with timely, data-driven crop intelligence across the entire growing cycle.
Background
Smallholder rice farmers in Nepal face significant crop losses due to late disease detection and limited field monitoring tools. This R&D project addresses that gap by integrating low-cost CCTV cameras, Raspberry Pi edge computing, and Google Cloud Platform into an automated, AI-driven pipeline — delivering continuous, scalable crop health monitoring across multiple paddy field sites.
Research Aim
To develop and validate a scalable, phenology-aware crop monitoring system that combines IoT edge computing, automated cloud pipelines, and AI-driven image analysis — enabling early disease detection, growth stage classification, and actionable agronomic insights for rice cultivation in Nepal's smallholder farming contexts.
