Deep Learning-based Lake Ice Detection using ESA Sentinel-1 SAR Data


09d78348 402a 4e27 9a61 39f64cdac999.jpg 500 500

Starting Date: earliest start: (2018-10-01) latest end: (2019-06-30)

Organization: Photogrammetry and Remote Sensing (Prof. Schindler)

Involved Host(s): Berger Monique

Abstract: Continuous monitoring of climate indicators is important for understanding the dynamics and trends of the climate system. Lake ice has been included in the list of Essential Climate Variables (ECVs). This thesis will investigate the possibility to monitor lake ice using deep learning with SAR data.

Description: This thesis will focus on an inter-disciplinary (intersection of artificial intelligence and climate change) research topic. Lake ice is an important variable to understand the regional and global climate change and has been recently recognized as an Essential Climate Variable (ECV). Monitoring and analyzing the (decreasing) trends in lake freezing using artificial intelligence techniques provides important information for climate research. Multi-temporal satellite images are a natural data source to survey ice on lakes. Unlike the optical satellite data, Synthetic Aperture Radar (SAR) is independent of data loss by clouds. For more background information, see http://www.prs.igp.ethz.ch/research/current_projects/integrated-monitoring-of-ice-swiss-lakes.html

Goal: The goal of this thesis is to monitor some target lakes (Sihl, Silvaplana, Sils, Aegeri, Greifen and St. Moritz) in Switzerland and to detect the extent and duration of ice, and in particular the ice on/off dates using SAR data. The major tasks are as follows: • Collection of ESA Sentinel-1 SAR data for Winter 2016/17 (October-April) and pre-processing. • Using the SAR data, perform spatio-temporal (estimate area and duration of frozen lake parts, multi-temporal analysis) monitoring of lake ice using Convolutional Neural Networks. • Precise estimation of the ice-on and ice-off dates. Note: Only computer vision / pattern recognition techniques are intended to be used (no knowledge on special interferometric / polarimetric radar data processing is required). Basic knowledge in Python / C++ and a deep learning framework (Keras, Tensorflow etc.) are required.

Contact Details: Manu Tom MSc. (manu.tom@geod.baug.ethz.ch)

Lake Ice Detection Convolutional Neural Networks Remote Sensing Computer Vision Texture Analysis Climate Change Satellite Image Processing Synthetic Aperture Radar


Labels: IDEA League Student Grant (IDL) Bachelor Thesis Master Thesis CLS Student Project (MPG ETH CLS)
Topics: Information, Computing and Communication Sciences Engineering and Technology Earth Sciences