unthinkable
Case study · Green City Watch · geospatial urban ecology

Planners in Jakarta see the city's lakes and rivers mapped from satellite images, with how stressed each one is.

Green City Watch, working with the Government of Indonesia and the World Bank, wanted Jakarta to protect its lakes and rivers from being filled, encroached and polluted.

What changed

95%accuracy in detecting surface water bodies (as reported)
90%precision (as reported)
11machine learning models trained (as reported)
What stood in the way

What stood in the way.

What stood in the way
  • Jakarta was losing urban lakes to filling, land use change, dumping and encroachment.
  • Finding each water body by season, measuring its sedimentation and tracking change over 35 years by hand was not practical.
What changed
  • 95%accuracy in detecting surface water bodies (as reported)
  • 90%precision (as reported)
  • 11machine learning models trained (as reported)
What we built

What we built.

  • A reliable data set from World View 1 satellite images, converted from TIFF to JPEG and cleaned of clouds and noise, then cut into grids.
  • Semantic segmentation models that label every pixel, to find water bodies and measure their stress.
  • Water boundaries found with the Normalized Difference Water Index, which separates water from dry land.
  • Semantic segmentation
  • NDWI
  • Satellite imagery
What it shows

What this work shows about us.

AI with a clear job

One question, where is the water and how healthy is it, answered pixel by pixel from images the client already had.

Data first

Most of the work was building a clean data set; the models are only as good as the images behind them.

Want to build similar satellite image AI for detecting water bodies? Tell us what you want to change. Our AI assistant, Ask Unthinkable, helps you define your requirements, then connects you with the right team.

Start your plan ↗
If we built it today

If we built it today, with AI.

What we would add now, moment by moment.

Monitoring

An agent would compare each new satellite pass with the last one and tell the planning team which lakes shrank or silted up, with the images to check.

Reporting

An agent would draft the quarterly note for the city on water body health from the model results, for an analyst to review.

Related

Where this work fits.

    Facts and figures from the project records; nothing added.

    Ask Unthinkable

    Want this for your business?

    Tell us what you run or what you're building. Our AI assistant, Ask Unthinkable, helps you define your requirements, then connects you with a senior AI engineer.

    Send us a note instead

    Required

    Thanks. A senior AI engineer will reply to you.
    u
    Ask UnthinkableClarity first, then the right team
    Your conversation

    An AI agent. This prototype uses prepared answers and sends nothing.