Course Description

Learn how to use SNAP to process multispectral data, namely Sentinel-2 MSI, to map land cover near the Tisza-Tó area, Hungary, in 2016.

Download Sentinel-2 MSI data, open the product on SNAP, and visualize it on RGB color composite. 

Use Sen2Cor algorithm to atmospherically correct Level-1C data and convert them to Level-2A. Learn the concept of RGB color composite and how it can be used for object identification. Resample and subset Sentinel-2 MSI image and then calculate various spectral indices (NDVI, NDWI, SAVI, etc..) and biophysical parameters (LAI and FVC). Apply k-means, Random Forest, and Maximum Likelihood classifiers on the region of interest, and then compare the accuracy of two supervised learning methods. Define Principal Component Analysis and apply it to reduce the dimensionality of processed multispectral data.

Course curriculum

  1. 01
  2. 02
    • Copernicus Sentinel-2 Mission

    • How to download Sentinel-2 imagery

  3. 03
  4. 04
    • Calculation of spectral indices

    • Biophysical processors

  5. 05
    • K-means classifier

    • Maximum Likelihood classifier

    • Random Forest classifier

    • Accuracy assessment and comparison

  6. 06
    • Demonstration and application

  7. 07
    • Course notes

    • Sentinel-2 imagery used in course

    • Vector data

  8. 08
    • Discussion on the course

Pricing - Life time Access

What will you learn?

  • Ηow to download Sentinel-2 MSI data.

  • Ηow to preprocess a Sentinel-2 MSI L1C product.

  • Ηow to calculate spectral indices and biophysical parameters.

  • How to apply unsupervised and supervised classifiers for land cover mapping.

  • Review how to compare random forest and maximum likelihood classifiers in terms of accuracy.

  • How to apply a features-reduction technique.

GEO Premium

Access our ENTIRE content instantly with a subscription

Student profile?

  • Undergraduate students

  • Master students and PhD candidates

  • Researchers and Academics

  • Professionals and Companies

About your Instructors

Ghada Sahbeni is a Ph.D. scholar in remote sensing for environmental modeling at Geophysics and Space Science Department, Eötvös Loránd University. She has studied Geology and then Agricultural Engineering with a major in Natural Resources Management at Carthage University. She holds a master’s degree in Environmental Engineering, specializing in air quality monitoring over urban areas at Pannonia University. The main objective of her scientific work is to apply remote sensing, GIS, and machine learning tools in soil salinity mapping over the Great Hungarian Plain. She has been involved in voluntary projects related to GIS, AI, and ML applications in various fields. Also, she is a former volunteer for the United Nations and currently a mentor in Women in Geospatial and African Women in GIS, professional networks to promote gender equality in the geospatial industry and academia. Moreover, she has supervised mentees and graduate students in Geographic Information Systems and remote sensing applied in environmental monitoring. Ghada is an editor at Maplines - British Cartographic Society and a member of the International Network on Salt-Affected Soils (INSAS) promoted by FAO.

Ghada Sahbeni

PhD Reseacher