Program
Master of Geographic Sciences – Remote Sensing
Advisor
Yacine Bouroubi
Field of research
Use of convolutional neural networks for object detection and classification of very high resolution satellite images
This project is in line with Professor Bouroubi’s GeoImageNet project to develop a very high resolution (THR) satellite imagery annotation web platform. The annotations collected on the platform will be used to train deep learning algorithms. These algorithms, once trained, allow the information to be extracted accurately and completely from THR images. At the end of the project, Canadian researchers will be able to use the annotated images and algorithms developed to achieve more innovations in the field of deep learning applied to remote sensing.