nspiration

We started working on VanSetu after looking at how cities like Delhi are growing and how much green space is being lost in the process. Rising temperatures, pollution and fewer connected green spaces are becoming serious problems. We wanted to explore whether technology could help cities deal with these issues in a more practical way.

We were particularly interested in satellite imagery and how much information it can provide about a city. We thought about using this data to understand which areas have less vegetation, which areas experience higher temperatures and where adding greenery could make a difference. This led us to the idea of VanSetu: a platform that could help identify potential green corridors and make urban greening decisions more informed.

What it does

VanSetu is a platform that uses satellite data and geospatial analysis to help identify and plan green corridors in cities.

It brings together different types of environmental data to give users a better understanding of a city's green cover, temperature patterns and environmental conditions.

Its main features include vegetation analysis using Sentinel-2 satellite imagery and NDVI, temperature analysis using MODIS Land Surface Temperature data, air quality monitoring using CPCB data, green corridor mapping, corridor prioritisation and interactive maps that display different environmental layers.

The main idea behind VanSetu is to go beyond simply mapping existing green spaces and help identify where new green spaces and connections are needed.

How we built it

We built VanSetu using Python, geospatial libraries and a React-based frontend.

We started by exploring different satellite and environmental datasets that could help us understand urban conditions. We used Sentinel-2 imagery for vegetation analysis, MODIS Land Surface Temperature data for identifying heat hotspots, NDVI for measuring vegetation cover and CPCB data for air quality information.

For geospatial processing, we used Python libraries such as GeoPandas, Rasterio, NumPy and Shapely to process geographical data, work with different spatial formats and perform spatial analysis. We also used Google Earth Engine for satellite-data processing.

The backend was developed in Python to handle data processing, geospatial operations and the analysis required by the platform. For the frontend, we used React, Vite and Tailwind CSS to build an interactive map-based interface where users can explore environmental layers and analysis results.

We deployed the frontend using Vercel and used GitHub for version control and collaboration.

Challenges we ran into

One of the biggest challenges was working with different environmental datasets. Satellite imagery, temperature data and air quality information all come in different formats and have different spatial resolutions.

We also faced challenges while combining datasets, processing raster and vector data, handling geographical information and identifying potential green corridors between existing green spaces.

Another challenge was making sure that the backend analysis and frontend map worked properly together. Some geospatial operations were computationally expensive, so we had to work on improving processing and keeping the platform responsive.

A lot of the work involved testing different approaches, fixing inconsistencies in the data and making sure the results made sense geographically.

Accomplishments that we're proud of

We are proud that we were able to bring together satellite imagery, environmental data and geospatial analysis into one platform.

We successfully developed an interactive platform focused on urban green corridor planning, integrated multiple environmental datasets and created geospatial workflows for vegetation and temperature analysis.

We also built map-based visualisations that make environmental information easier to explore and understand.

One of our biggest achievements was securing 2nd place at IG:NITE under Team Coffee Bean for VanSetu.

What we learned

Working on VanSetu helped us understand how challenging it can be to work with real-world geographical and environmental data. We learned how satellite imagery can be used to study cities and how different datasets can be combined to understand environmental problems.

We also gained practical experience with Python-based geospatial tools, remote sensing, frontend and backend development, and deploying a complete web application.

One of our biggest takeaways was that building a useful product is not just about getting the technology to work. It is also about understanding the problem, making the data meaningful and presenting the results in a way that people can actually use.

What's next

We want to continue improving VanSetu by adding more environmental datasets, improving green corridor identification and prioritisation, and making the analysis more useful for actual urban planning.

We also want to explore predictive features that could help users understand how different greening interventions might affect urban temperatures and environmental conditions over time.

Our goal is to make VanSetu a useful tool for people working towards greener and more sustainable cities.

Built With

  • climate
  • data-visualization
  • environmental-analytics
  • fastapi
  • geopandas
  • geospatial-analytics
  • gis
  • github
  • google-earth-engine
  • javascript
  • modis
  • ndvi
  • numpy
  • python
  • rasterio
  • react
  • remote-sensing
  • rest-apis
  • satellite-imagery
  • sentinel-2
  • shapely
  • tailwind-css
  • vercel
  • vite
Share this project:

Updates

Submission history