Learn how to use Python to access the Copernicus Dataspace Ecosystem, and process and analyze Sentinel-2 imagery
The use of remote sensing data is growing, with the need to use such data for many applications ranging from the environment to agriculture, urban development, security and disaster management. This course is intended for beginners who would like to make their first acquaintance with remote sensing data, and learn how to use freely available tools such as Python to analyze and process freely available imagery from the Copernicus Sentinel-2 mission. No prerequisite knowledge is required.
What you’ll learn
- Create a Copernicus Open Dataspace account.
- Install and setup Anaconda for Python development.
- Search, filter and download Copernicus Sentinel-2 data using the Python API.
- Analyze and process Copernicus Sentinel-2 data.
Course Content
- Introduction –> 3 lectures • 12min.
- Downloading Copernicus data –> 3 lectures • 13min.
- Opening and processing a Sentinel-2 acquisition –> 4 lectures • 15min.
- Commonly used indices: NDVI and NDWI –> 2 lectures • 8min.
- Exporting data –> 1 lecture • 3min.
- Additional content: land cover mapping –> 1 lecture • 6min.
- Conclusion –> 1 lecture • 1min.
Requirements
The use of remote sensing data is growing, with the need to use such data for many applications ranging from the environment to agriculture, urban development, security and disaster management. This course is intended for beginners who would like to make their first acquaintance with remote sensing data, and learn how to use freely available tools such as Python to analyze and process freely available imagery from the Copernicus Sentinel-2 mission. No prerequisite knowledge is required.
Through a step-by-step learning process, this course starts off with setting up a Copernicus Dataspace Ecosystem account, and installing a Python environment. Python is then used to make use of the Copernicus Dataspace Ecosystem API to search for, filter and download Sentinel-2 products. Also using Python, these products are then opened and the corresponding optical and near-infrared bands are analyzed and processed to create and RGB composite image, as well as calculate commonly used indices such as NDVI and NDWI. Basic correction methods such as normalization and brightness correction are also introduced.
At the end of the course, a bonus application is presented, where a machine learning technique (clustering) is used to partition the content of the Sentinel-2 product into various categories to obtain an estimate for a land cover map.