WARP

An AI platform that helps governments prevent groundwater pollution and recover value from dumpyards using drone and climate intelligence.

Comment

WARP — Waste Analysis and Report Generation Project

Inspiration

Unscientific waste dumping and its long-term impact on groundwater is a serious but often unnoticed environmental issue. Many dumpyards operate without continuous monitoring, and groundwater contamination is usually detected only after irreversible damage occurs.

WARP was inspired by the idea of using AI, drone imagery, and verified government data to support early-stage environmental risk assessment and informed decision-making.

What it does

WARP (Waste Analysis and Report Generation Project) is an AI-assisted system that:

  • Accepts drone images of dumpyards
  • Analyzes surface-level waste distribution and dumpyard spread using computer vision
  • Integrates official groundwater data from government sources
  • By analysing the ground water level the weather since the past years and predict when the groundwater level is effected by the seepage caused by the dumpyard
  • Automatically generates a structured environmental assessment report
  • Sends the report to relevant stakeholders via email

The system acts as a decision-support tool and does not replace physical inspection or laboratory testing.

How we built it

Frontend

  • HTML (72.0%) – application structure
  • CSS (11.7%) – styling and layout
  • JavaScript (10.9%) – client-side logic and API handling

Backend & Intelligence

  • Python (5.4%) – image processing and report generation
  • Gemini API – intelligent analysis and summarization
  • Official groundwater data from the NITI Aayog website
  • EmailJS – automated email delivery

Workflow

  1. Upload drone images of a dumpyard
  2. Analyze surface-level waste patterns
  3. Retrieve regional groundwater data
  4. Generate a structured PDF report
  5. Email the report to authorities

Challenges we ran into

  • Drone imagery provides only surface-level information
  • Correlating visual waste data with groundwater statistics required careful assumptions
  • Avoiding scientifically unsupported claims without physical sensors
  • Limited time prevented implementation of thermal imaging and 3D volumetric analysis

Accomplishments that we're proud of

  • Built an end-to-end automated reporting system
  • Integrated AI with verified government datasets
  • Maintained transparency about technical limitations
  • Designed a solution suitable for policy and planning use

What we learned

  • Environmental AI systems must be scientifically responsible
  • Government data significantly improves system credibility
  • AI works best as a decision-support mechanism
  • Clear communication of limitations builds trust

What's next for WARP

Future Scope

  • Thermal and hyperspectral drone analysis
  • 3D dumpyard volume estimation using photogrammetry
  • Leachate seepage and groundwater risk prediction models
  • Vegetation stress analysis around dumpyards
  • Landfill mining feasibility assessment
  • Selling dumpyard waste to authorized recycling companies
  • B2B marketplace and reverse auction mechanisms

These additions will transform WARP into a predictive environmental intelligence and waste-to-value platform.

Built With

Share this project:

Updates