I’m experimenting with feature detection and matching in OpenCV. I have the flight data corresponding to each of these images so I can also estimate a lon/lat/orientation for each keypoint. Then I’m thinking how I can find these features in additional images and try to come up with a best fit/consensus among a large pile of images in a geo-consistent/aware way. I know there are a few other tools to do some of this, but I haven’t found anything open source that really handles aerial imagery taken at different angles so I’ve started my own experimentation. Does anyone know of tools that can do this stuff already? Maybe I missed something obvious … As much fun as this stuff is, I’d hate to reinvent the wheel if I don’t have to.
Look for mavmap on github. Best regards