The Department of Earth Observation Science (EOS) of the Faculty of Geo-Information Science and Earth Observation (ITC), University of Twente in Enschede, the Netherlands, has a full-time vacant position for a PhD Candidate.
Your tasksMaps are quickly becoming outdated, about 10 % of the objects change annually. This project focuses on methods to automatically interpret newly acquired sensor data to detect and to update the changed objects in the digital map. The sensor data includes high resolution 2D aerial image data and 3D laser scanner data. Our approach is to use the existing (old) digital maps to learn how various objects appear in these 2D and 3D datasets. It is your task to design a deep learning network to semantically segment airborne sensor data, making use of a huge training dataset derived from the fusion of map and sensor data. For (3D) point clouds derived from stereo (2D) images we aim to build a combined 3D and 2D classification approach, to be able to later classify both point clouds and imagery. Alongside this PhD position, one Postdoc will work on the generation of training data, and one other PhD Candidate will work on the delineation of data into map objects.
The specific tasks in this project are:
- Design of neural networks for image and point cloud classification using massive training data
- Image and point cloud classification and domain adaptation
- Change detection between sensor data and map data