Back to projects

Flagship research · Agriculture

SPADE-AI

Smartphone-based soil analysis and crop-planning intelligence

Funded · UGC, ADBRunning · 2026–2027
  • Agricultural AI
  • Computer Vision
  • Soil Science

Developing an accessible system that estimates soil properties from smartphone images and combines them with climate and geospatial data to support Bangladeshi farmers.

How it works

  1. Step 1

    Field capture

    Farmers photograph soil samples with an ordinary smartphone in the field.

  2. Step 2

    Spectral reconstruction

    Deep-learning models estimate hyperspectral signatures and key soil properties.

  3. Step 3

    Agro guidance

    Fused climate and geospatial data deliver crop suitability and rotation advice.

Research objectives

  • Estimate key soil properties — including nitrogen, phosphorus, potassium, pH, organic matter, and micronutrients — from ordinary smartphone images.
  • Collect geo-referenced soil samples with RGB and hyperspectral imagery from agro-ecological zones across Bangladesh.
  • Integrate soil, climate, geospatial, and crop data to provide location-specific crop suitability and rotation guidance.
  • Reduce dependence on costly laboratory testing and specialized hyperspectral equipment for precision agriculture.

Methodology

  1. 01

    Collect around 1,000 soil samples from approximately 10 agro-ecological zones.

  2. 02

    Develop deep-learning models for RGB-to-hyperspectral reconstruction and soil-property prediction.

  3. 03

    Fuse reconstructed spectral information with multi-source agro-ecological, climatic, and geospatial datasets.

  4. 04

    Validate predictions against laboratory measurements and field observations.

Expected outputs

  • Deep-learning models for spectral reconstruction and soil-property estimation.
  • A Bengali-language mobile application for farmers to capture soil images and receive guidance.
  • Crop suitability and rotation recommendations grounded in local soil and climate conditions.
  • Open research outputs supporting sustainable agriculture and improved rural livelihoods.

Publications and datasets

Research papers, models, and open datasets will be shared here as they are released.

Project updates

Milestones, fieldwork notes, and lab announcements will appear on our notices page.

View notices →