I am trying to create a 3D model of a sea wall using laser scans and photography. The scans were acquired with a terrestrial Topcon scanner, registered only a local grid system and exported with colour as .e57 files. My workflow, using one scan and a series of nearby images is as follows:
-
Import one scan. Tick registered/georeferenced=TRUE and confirm that COLOUR is present Import the images
-
Run alignment. Two components result. Now, this is the problem part. I want to merge the two components. This does not happen automatically, so I add four control points to two of the images. The XYZ coordinates of these control points are taken from the pointcloud, so are on the same coordinate system.
-
Re-run alignment. I find it impossible to merge the two components and end up with one good model that is a hybrid of both the laser scan and images. HELP!
Possible issues: I have read that the software uses the colour to match the laser scan and images. In the case of this site, there is very little colour (it is a grey Limestone wall) so I wonder if that would adversely affect it?
The images from the camera are geotagged with lat/long. Could this be confusing things? I did change Absolute Pose to “unknown” that removed the EXIF position from the list,
I have tried the setting camera priors=false I have read all the forum posts and watched all the videos that, when it works, makes this look very easy. I think my case is a very common one, so it would be good to see a list of all the alignment settings to get it to work. Thanks in advance. Nick