Advancing Remote Sensing based Machine Learning for Predictive Wildlife Land Use Classification
(Working Title)

The overall goal and guiding theme of the dissertation is the extraction of semantic information from remote sensing data by using and developing novel methods of computer vision and machine learning. A special focus lies on the contribution to a greater understanding of the presence predictions of larger mammals and the monitoring of their specific habitats.

Group of Prof. Dr. Denis Jordan, Chair for Applied Mathematics and Statistics at the Institute Geomatics, FHNW (University of Applied Sciences Northwestern Switzerland)

Research Collaboration at UZH/RSL with Dr. Mathias Kneubühler and Prof. Dr. Alexander Damm

• Remote Sensing
• Image Analysis
• Object Detection
• Computer Vision
• Deep Learning
• Data Mining
• GIS
• Land Use / Land Cover Classification
• Ecological Biogeography
• Animal Demography
  
• Human-Wildlife Conflict Mitigation
• Thermography
• UAVs / Drones