Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Topics - JMerc

Pages: [1]
Hello all,
I am attempting to use aerial photos (with high accuracy and precision camera locations) to extract surface elevation information without GCPs. There is about a 6 photo overlap between the photos in the area I am interested in (4-band imagery, including an infrared band). The ground resolution of the pixels is 30 cm. The camera used is a an UltraCamXp. A camera-specific calibration model has been created for the camera (see attached).

The problem is that I am at a loss as to how to convert the parameters within the provided documentation into something usable by PhotoScan. I've tried using what seem to be standard equations (e.g., some of the equations provided here - to try and relate the given and desired parameters, but when I finally am able to run the model (after double and triple checking that the values make sense relative to each other), the output is garbage; it looks way better without the camera calibration. However, when I compare to validation points, the updated camera positions can be as much as 120 m off, elevation wise, and between 100-200 m off, horizontally. Once I re-run the model with GCPs, those values are off 10, and 15-25 m, respectively, though the whole point of my exercise is the not have to use GCPs.

I'm thus wondering if anyone has any ideas concerning a program or exact formulas one might use to perform an appropriate conversion related to fx, fy, cx, cy, k1, k2, and k3 given the documentation I've uploaded. I don't have access to the original camera, so I can't calibrate it using Lens.

Thank you for your time,


Pages: [1]