Skip to main content
AboutProblemsGuidelinesTerms & conditions

Problem Statement 3

"Satellite-Based Crop Health & Resource Advisory System"​

Focus: Satellite Data Processing | Vegetation Monitoring | Temporal Analysis | Geospatial Insights

1. Background:​

Free satellite data (Sentinel-2, Landsat, MODIS) provides 10m-resolution imagery globally—a resource for monitoring crop health. However, farmers can't interpret raw satellite data, and even with available imagery, three critical problems remain unsolved.

2. Current Problem:​

1. Data Accessibility Gap

  • Satellite imagery exists but requires GIS expertise to use
  • Farmers can't translate NDVI/NDWI values into actionable insights
  • Impact: Valuable monitoring data stays unused

2. Manual Inspection Limitations

  • Field surveys are slow, expensive, and don't scale
  • Crop stress detected only after visible damage
  • Impact: Late interventions, reduced yields

3. No Trend Monitoring

  • Single-point observations miss seasonal patterns
  • Can't compare current season vs historical averages
  • Impact: Miss early warning signs of drought or stress

3. Goal:​

Build a system that converts satellite imagery into simple crop health advisories.

Transform raw data → Create a platform that:

  1. Processes satellite imagery to compute health indices
  2. Analyzes temporal trends (weekly/monthly)
  3. Generates interpretable insights (not just maps)
  4. Visualizes crop conditions through farmer-friendly dashboards

4. Expected Solution (MVP Requirements):​

1. Satellite Data Pipeline​

Auto-fetch and process imagery:

  • Data Sources: Sentinel-2, Landsat-8, MODIS
  • Preprocessing: Cloud removal, resampling
  • Coverage: District/block-level analysis (no farm-level shapefiles needed)

Output: Clean, analysis-ready satellite imagery

2. Health Index Computation​

Calculate key indicators:

  • NDVI: Crop vigor, growth monitoring
  • NDWI: Water stress detection
  • LST: Heat stress identification
  • Anomaly Detection: Current vs historical comparison

Output: Numerical scores + color-coded severity (healthy/stressed/critical)

3. Temporal Trend Analysis​

Track changes over time:

  • Weekly/Monthly Trends: Growth progression monitoring
  • Seasonal Comparison: 2024 vs 2023 same period
  • Alert Generation: Flag sudden drops in vegetation health (>20% decline)

Validation: Show a real crop stress event detected 2-3 weeks before field reports

4. Advisory Dashboard​

Real-time visualizations:

  • Health Heatmap: Color-coded crop condition maps
  • Trend Charts: NDVI/NDWI changes over 90 days
  • Alert Panel: Active stress zones with severity
  • Impact Metrics:
    • Area monitored (hectares)
    • Stress zones identified
    • Temporal coverage (weeks of data)

5. "Level Up" Features:​

Advanced Features (Choose 2-3)

  • Drought Forecasting: Predict water stress 2-4 weeks ahead using trends
  • Crop Type Classification: Auto-identify rice/wheat/cotton from imagery
  • Radar Integration: Use Sentinel-1 for cloud-independent monitoring
  • Mobile Alerts: SMS notifications for detected crop stress
  • Multi-Crop Support: Handle 5+ crop types with specific thresholds

6. Tech Stack:​

Core

  • Backend: Node.js/Express + Python/FastAPI (geospatial processing)
  • Database: MongoDB (metadata) + PostGIS (spatial data)
  • Storage: Cloud storage for satellite imagery

Geospatial

  • Processing: Google Earth Engine API or rasterio/GDAL
  • Analysis: NumPy, SciPy (index computation)
  • Visualization: Leaflet.js, Deck.gl (maps)

Frontend

  • Framework: React.js
  • Charts: Recharts (trend analysis)

External APIs

  • Copernicus Dataspace (Sentinel), USGS EarthExplorer (Landsat), NASA LPDAAC (MODIS)