We will be undergoing planned maintenance on January 16th, 2026 at 1:00pm UTC. Please make sure to save your work.

Constant

An easy-to-use platform made for change detection through the SAR imagery dataset provided by Ursa Space System on the AWS data exchange platform.

Comment

Inspiration

As the saying goes, a picture is worth a thousand words; in the case of satellite imagery, considerably more so. However, sometimes these pictures are worth much less or nothing at all since traditional imagery faces capture obstacles such as cloud cover and nightfall. These limitations have been taken care of by Synthetic Aperture Radar (SAR) satellites. SAR satellites are designed for change detection. Images of the ground taken from the same location in space will always be identical unless something changes.

Need of monitoring the change:

  1. Agriculture. Differences in surface roughness are indicative of field plowing, soil tillage, and crop harvesting.
  2. Floods. Differences in surface reflection can help distinguish heavy flooding, light flooding, urban areas, and permanent bodies of water.
  3. Land subsidence. Differences in measurements over time can reveal displacements of land, such as sinking ground caused by the extraction of underground natural resources.
  4. Snow cover. Differences in surface reflection can help forecast snowmelt by distinguishing wet snow, dry snow, and snow-free areas.
  5. Wildfires. Penetration through thick smoke can provide more accurate and timely information about the extent of a forest fire and can help quantify vegetation loss.
  6. Wetlands. Penetration through wetland areas can reveal flooded vegetation where land is covered by shallow water.

Therefore, we have thought of a solution that can provide easy to use interface for researchers and enthusiasts that can monitor historical changes around the globe without any hassle to look at millions of historical images.

What it does

Constant is a web application, that is made for change detection. It provides an easy-to-use platform for users to search through the map interface for SAR images based on geo-locations and timestamps according to the need and use case. It looks through the world's largest network of commercial SAR satellites that has an archive of over 10 million images and 14 years of history. And present all the images in a sorted manner right on the top of the map in order to provide the user whole some idea about the images such that from where, what time, and how these images were taken by the SAR satellites.

How we built it

Backend: This application is powered by the AWS Data Exchange API: SAR Virtual Constellation: Catalog from Ursa Space Systems. It consists of 10 million images and 14 years of history, collected and maintained widely from the different satellite vendors. In this application, we have utilized the data from ICEYE with the image acquisition in Spotlight Mode. Although we have the flexibility to look after the location in multiple modes, spot mode enables the finest resolution and largest available scene size for detailed monitoring, delivered in any weather.

These are the parameters we have considered while requesting data from the datasets:

  1. Acquisition Mode: Spotlight
  2. Vendor: ICEYE
  3. Satellite Pass: DESCENDING

We have developed a lambda function, that provides the REST API gateway for the client application to interact with the data safely and according to the need. With the help of the query parameters, the request GraphQL query to access data is made dynamic such that we can query data based on a specific location or/and specific date window.

Client Interface: The frontend web application is developed with the javascript framework ‘React’. It utilized the Open Street Map implementation of the global map in order to provide a hassle-free user experience. We have used other node packages to ensure fast development.

Deployment: Deployment of the compliments is done over the AWS cloud. The client-side application is deployed on the SAAS service Amplify and the backend API is deployed as a serverless HTTP-triggered lambda function communicating with the client application through API Gateway.

Deployment

Challenges we ran into

The major challenge with this hackathon was the improper documentation of the different datasets. It was somewhat difficult to figure out what kind of data APIs provide. Also, as we need to raise the request to subscribe to the dataset until the fulfillment of the request we can’t explore the data and develop ideas around that.

Accomplishments that we're proud of

After facing multiple issues with different datasets and thinking about multiple ideas, We are glad we are able to pull off something from this hackathon. From understanding the AWS Exchange Dataset to deploying the final bit of code, we have been able to complete this project with my hectic work schedule.

What we learned

This is the first time we have worked with the AWS Exchange Datasets, it was great learning to know how to use dataset endpoints in real-life applications. Along with that, we had to learn about SAR imagery, its fundamentals, and its applications.

What's next for Constant

With the interest of time available for the submission, we were able to focus on only one vendor i,e ICEYE of the URSA space systems. In the future, we can figure out the similarities between the data provided by the different vendors and provide the best results to the users.

Share this project:

Updates