Detecting deforestation using unsupervised K-means clustering on Sentinel-2 satellite imagery and SageMaker Studio Lab(SMSL)
A Jupyter Notebook that connects to Amazon Sustainability Data Initiative (ASDI) datasets from sentinel hub to show classification to show deforestation.
Human-driven and natural loss of trees—deforestation—affects wildlife, ecosystems, weather patterns, and even the climate.
Deforestation around the world has reached a critical level, causing irreversible damage to environmental sustainability that is contributing to climate change around the world. Widespread forest fires, from the Amazon Basin in Brazil, to the west coast of the United States, are raging all year-round. This notebook will allow us to detect deforested areas in california wild fire using satellite imagery.
ASDI Homepage: https://sustainability.aboutamazon.com/environment/the-cloud/asdi
ASDI Datasets: https://registry.opendata.aws/collab/asdi/
You can sign up for SageMaker Studio Lab and use it for free without an AWS account. You can run for 4 hours with GPU or 12 hours with CPU and then logout and log back in for another session. Your data and notebooks are persisted. After clicking the launch button below, choose "download whole repo".
When it's done installing and configuring the environment, open the .ipynb notebook file. Click-Enter to run each row and wait a moment to see the results of each line before proceeding to the next. The line marker should change to a number when it's successfully run that line, ie "[5]" means that it has run line 5.
See CONTRIBUTING for more information.
This library is licensed under the MIT-0 License. See the LICENSE file.