This training is in the past.
Introduction to SAR data and Application
Day 1 – Land cover/ Crop type classification
Morning: During the first morning session, a brief introduction to satellite remote sensing will be given, with emphasis on the different information content and complementarity of SAR (Sentinel-1) and optical data (Sentinel-2). The second session will focus on feature extraction and introduce different types of classification algorithms (e.g. Bayisian, K-means, Random forest). Their pros and cons will be presented with special attention to multi-waveband data and multi-temporal data. Concerning SAR, also the inherent speckle noise and its influence on classification will be covered.
Afternoon: Hands-on exercise within a RUS environment. In this tutorial, we aim to introduce Optical (Sentinel-2) and SAR (Sentinel-1) data feature level fusion for land cover and crop monitoring on a study area in Denmark. We will use a time series of optical and SAR GRD data to demonstrate how multi-temporal and multi-sensor approach can increase the accuracy of our classification. Optical data provide accurate land cover/crop classifications if the cloud free acquisitions time series is large enough and if the acquisition dates are taken in the suitable period of the year. However, in study areas with extensive cloud cover, the number of available optical images could be insufficient for accurate classification. SAR time series from Sentinel-1 can be used combined with the optical Sentinel-2 data in these cases to improve the results.
DAY 2 – Sentinel-1 InSAR processing for ground motion monitoring
Morning: Two theoretical lectures will be given. The first will provide a top-down introduction to SAR interferometry (InSAR), with emphasis on its application to the measurement of surface deformation. The working principles, applicability requirements, and main error sources of InSAR techniques will be presented starting from examples and covering the relevant theoretical concepts. The second lecture will address specific properties of the Sentinel-1 SAR Terrain Observation with Progressive Scans (TOPS) acquisition mode, and discuss the processing steps required to carry out InSAR deformation measurements.
Attendees must bring their own laptop and will be provided with access to the RUS platform Virtual Machine.
Attendance to the training course is free of charge, however travel and accommodation expenses are excluded.
Audience & Selection
- This training course is addressed to participants with basic knowledge of remote sensing
- Participants will be selected to attend on the basis of the questionnaire filled in at application
- Confirmation e-mails to selected attendees will be sent in early November 2018
The Training is divided on both days in two sessions, morning and afternoon.
13 December → 9:00 – 17:00
14 December → 9:00 – 17:00
VenueDTU Lyngby Campus, building 303A, Auditorium 49. More info on how to reach the campus here.
Two coffee breaks will be served each day in the venue facilities.
Max number of participants
Thursday 13th of December 2018 - Friday 14th of December 2018
Asmussens Allé, 2800 Kongens Lyngby, Denmark