AI for Agriculture (AgriTech)

Course outline and learning expectations

This is a tutor-led course. The outline shows what your tutor will cover; teaching materials and examinations are provided directly to enrolled students.

What you will learn

• Explain practical AI applications and limitations in agriculture in Uganda, East Africa and Africa
• Prepare and explore crop image, weather, soil and farm datasets
• Build and evaluate an image-classification model for crop disease screening support
• Develop a baseline yield-prediction model and interpret its performance
• Design pest and weather-risk alert rules with clear confidence and escalation limits
• Calculate and interpret NDVI for basic crop-health monitoring
• Translate precision-agriculture concepts to smallholder farming contexts
• Prototype an accessible farmer-facing SMS/USSD workflow
• Apply responsible AI practices covering data quality, bias, privacy, consent and human oversight
• Build and present a locally relevant AgriTech prototype

Modules

1. 1

   AI in African agriculture

   Evaluate practical uses, limitations and adoption conditions for AI in agriculture, with emphasis on Uganda, East Africa and African smallholder systems

   AI and digital agriculture landscapeAgricultural value-chain problemsUse-case selectionSmallholder constraintsResponsible AI and human oversight
2. 2

   Crop disease detection with image classification

   Prepare crop-image data and build a baseline classifier for disease-screening support using locally relevant crops

   Image collection and labellingData quality and class balanceImage preprocessing and augmentationTransfer learningAccuracy, precision, recall and confusion matricesField validation limits
3. 3

   Yield prediction using weather and soil data

   Create and assess baseline models that estimate crop yield from weather, soil and farm-management variables

   Data cleaning and feature engineeringWeather and soil variablesRegression modelsTrain-test validationError metricsInterpretation and uncertainty
4. 4

   Pest and weather-risk alerting systems

   Design practical early-warning workflows that turn forecast and field data into clear, actionable alerts

   Risk indicators and thresholdsWeather API or supplied dataPest surveillance dataFalse alarms and missed alertsEscalation to extension expertsLocal-language message design
5. 5

   Precision agriculture for smallholder farming

   Adapt precision-agriculture principles to affordable, small-scale and cooperative farming settings in Africa

   Site-specific managementFarm mapping and samplingInput optimisationLow-cost sensors and mobile dataCost-benefit and inclusionFarmer-centred design
6. 6

   Satellite and remote sensing data with NDVI

   Use basic satellite imagery and NDVI to observe vegetation condition and discuss limits such as cloud cover and spatial resolution

   Remote sensing fundamentalsSatellite data sourcesRed and near-infrared bandsNDVI calculationTime-series interpretationGround-truthing
7. 7

   Farmer-facing SMS/USSD tool

   Prototype a low-bandwidth service that delivers simple recommendations or alerts to farmers and extension workers

   User research and service flowUSSD menus and SMS messagesAPI sandbox integrationConsent and data protectionError handlingUsability testing
8. 8

   Capstone AgriTech prototype

   Build and present a focused prototype, such as a coffee crop-disease screening tool linked to an SMS alert workflow

   Problem definitionDataset and model cardPrototype developmentTesting with representative usersDeployment considerationsCapstone demonstration