NamadaIsabelle/shamba-assist-app

# ShambaAssist 🌱 > AI-powered crop disease detection and soil health monitoring for smallholder farmers in Kenya. ShambaAssist combines IoT sensor data with machine learning image analysis to give farmers early warnings, actionable insights, and practical recommendations — designed from the ground up for low-bandwidth environments. --- ## The problem Smallholder farmers in Kenya lose significant yields every season to crop diseases and poor soil health monitoring. Existing tools are either too expensive, require reliable internet, or aren't trained on East African crop varieties. ShambaAssist is built specifically for this gap. --- ## How it works A farmer is onboarded by an agronomist who sets up sensors on their plot. Based on farm size, a combination of real ESP32 sensors and simulated sensors cover the field. Sensor data flows through an MQTT pipeline to the backend, where it is combined with crop images uploaded by the farmer or agronomist. The AI layer fuses both data sources to generate a diagnosis and recommendation, which is pushed to the farmer's dashboard and delivered via SMS for areas with no internet. Farmers who want in-person support can request a physical visit from their assigned agronomist directly through the app. --- ## Features - **Sensor fusion** — soil moisture, NPK, temperature and humidity combined into a unified crop health score - **Crop image analysis** — ML model trained on East African crop varieties detects leaf disease, root conditions and fruit ripening - **Hybrid IoT deployment** — real ESP32 sensors anchored by Wokwi-simulated sensors to cover larger plots affordably - **SMS alerts** — critical warnings delivered via Africa's Talking API for low-bandwidth areas - **Agronomist dashboard** — manage assigned farmers, review diagnoses, monitor sensor health, handle visit requests - **Farmer dashboard** — view crop health, sensor readings, AI recommendations and nearby agronomists - **Visit request system** — farmers can re …