This code pattern demonstrates how to analyze satellite data to derive useful information. It uses a real-world use case that can be used as a template to analyze any satellite data.
Satellite data is becoming more and more popular in the recent years, thanks to more accessible high-resolution satellite products which contain vast amount of information. In this Code Pattern, we will utilize a real world use case to demonstrate how to retrieve useful information and insights from satellite data. Most components of this demo can be generalized and used towards any satellite data. Therefore, you can use this code pattern as a template to analyze your own satellite data.
With the native spatiotemporal capabilities of Watson Studio, analyzing satellite data is very straightforward. The demo provides a set of key features that can be used for any satellite data regarding how to:
- Align and join multiple satellite layers in a generalized way
- Apply machine learning on satellite data
- Use the Watson Studio native spatiotemporal capability to transform and manipulate satellite data
The intended audience for this code pattern is satellite-domain developers and users who want to learn how to analyze satellite data efficiently.
- Load “Vectorized Dynamic Surface Water Extent” data from the Data Asset eXchange (DAX).
- Align and join two layers from the data to generate a derived layer.
- Apply machine learning on the derived layer to improve the accuracy.
- Apply spatiotemporal analysis and transformations to further improve the accuracy and derive insight.
- Visualize the results from each step.
Get detailed instructions in the readme file. Those steps tell you how to:
- Sign up for Watson Studio.
- Create the notebook.
- Run the notebook.