This project provides a procedure and (python) scripts to make it (relatively) easy to
- create a training dataset for a topic of your choice
- train a neural network to segment orthophotos
- run the segmentation on a larger area + vectorize the result
- apply some basic postprocessing like dissolve,...
Only open source software is needed, eg. QGIS and tensorflow.
Installation and usage instructions can be found in the orthoseg wiki