Mixing camera none-geopositioned images with drone geopositioned images.
Hello,
I have about 900 (61 megapixel) images of a construction site that was taking with Sony A7 IVR camera. The generated photogrammetry model looks very good but have few missed spots. Also these images do not have GPS data in it. I did not apply any scaling or control points to the photogrammetry model. I also have another 300 images that was taking with drone 12 megapixel camera from about 16 feet above the site. Drone images do have GPS data. Can I improve original photogrammetry model by adding images taking with the drone? Is this is as simple as just pointing RC to the another folder with images from drone? Will added images (with GPS data) will automatically bring a correct scale into the resulting model? Will added images produce a correctly geopositioned model?
-
Hi Neugroza,
yes, generally it is the same as adding another set of the images.
These datasets will align automatically, if there is a good overlap between datasets. You can also follow this workflow to merge these kind of datasets: https://www.youtube.com/watch?v=kKA-KxT25KA. Also, there are another tutorial on your YouTube channel, which can help you in this case.
GPs data from drones aren't so precise, but for some basic georeferencing it is good enough. It is always better to use some scale during capturing or some ground control points with measured coordinates to achieve better precision and position of the created model. But it also depends, what is a purpose of your model.
Please sign in to leave a comment.
Comments
1 comment