Satellite Data Analysis in Python

Module code: GY7709

This module focuses on the Python programme language applied to the analysis of satellite data, particularly from the Copernicus Sentinel missions. Python is increasingly sought after by employers in the space industry and allows the efficient use of cloud computing and high-performance computing to process large amounts of satellite imagery. In this module, you will learn how the Python programme language can be used to process satellite images, including automated data searches, pre-processing, visualisation, image enhancement and applications. You will learn how to use machine learning to derive value-added data products and services from Earth observation data and how to communicate your results.

The lectures will cover the principles of environmental remote sensing, properties of satellite image data, image processing and transformations, image enhancement and visualisation, geographic projections, the design of new Earth observation applications and services, as well as time-series analysis, image classification and machine learning (random forests and maximum entropy).

The practical sessions will focus on the automation of data searches, data downloads, pre-processing (incl. atmospheric correction), changing map projections, using Python on cloud or high-performance computing, time-series analysis for spatial data stacks, as well as land cover classification using maximum likelihood, random forest and maximum entropy.

Learning

  • 10 hours of lectures
  • 10 hours of practicals
  • 10 hours of supervised lab time
  • 120 hours of independent study

Assessment

  • Practical portfolio (30%)
  • Oral presentation of two selected journal papers (10%)
  • Journal paper style project report (60%)