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.

Messages - fsteiner

Pages: [1]
Hejhej there,

I would like to compare LPS and PhotoScan calibration results. I need to correctly interpret the estimated camera calibration parameters. I assume the same as stated by nadar here:

Then, I need to introduce the estimated camera calibration parameters.
Here also, LPS is quite confusing:
I need focal length (this one is easy) and x&y decentering offsets in mm, while photoscan reports these in pixels. I assume the offset is given by the difference between the actual center and the theoretical center (= Xsizs / 2).  It's easy to transform by multypling the offset in pixels by the pixel size, but I'm not sure about the direction (sign) of this parameter. Could somebody confirm if [Cx-(Xresolution/2)]*pixelsize corresponds to Xoffset of principal point, as requested in LPS

But then, the sign is obviously the wrong way (in x and y direction). It could be because of a different calibration result, but I rather think, something is "upside down" (see attachment). Any ideas, what I did wrong? Couldn't find anything on the definition of +/- for LPS, PS is more clear I think.

best regards
- plitvicer

General / Re: Question regarding pixel size
« on: June 29, 2015, 07:16:44 PM »
Thank you for this answer,

Just to make sure I get this right: If the pixel size is used for the initial calibration estimation, it wouldn't improve anything, when adding these values in the end? Or in other words: If the initial alignment and the camera calibration work fine, the pixel size/pitch is not necessary?

Thank you very much,
- fsteiner

General / Question regarding pixel size
« on: June 25, 2015, 06:21:24 PM »
Hello community,

in the camera calibration settings, I can set the pixel size (mm). I have two questions regarding this vaule:
1. Which pixel size is of interest? Many cameras have a higher density of pixels on their actual sensors than the resulting pictures will have (effevtive vs. total). Which value should be used?
2. At which point in the calculation is this value used? What does it change, when adding the value in the end? Which steps need to be redone?

Best regards.

Bug Reports / Re: Circles in DEM
« on: March 26, 2015, 06:08:34 PM »
Thank you for your answer.

I tried different settings for dense point cloud generation, the artifacts always appeared. I used the highest quality for the example picture I posted, but I also tried lower qualities.

What do you mean by "alignment issues affecting the dense cloud already"? Is there a way to improve the alignment, concerning these artifacts?


Bug Reports / Circles in DEM
« on: March 26, 2015, 04:02:44 PM »

I am using PhotoScan to generate a Digital Elevation Model. I have almost 60 pictues, taken from a UAV platform, the coverage ist pretty well and I get a final result with a ground resolution of 2,2 cm.

I chose the following setting:
- Build Dense Cloud: Very high
- Build Mesh: highest Option; no interpolation
- Export DEM: suggested resolution (2,2cm)

The DEM looks normal, but when I use the DEM to calculate a shaded relief in ArcGIS, I can see many circles in some areas. They definitely come from the PhotoScan-model. My first idea was, that they might be because of the interpolation during mesh building, but even if turned off, they are still there. I tried several other options (resolution, quality-settings), but the result always shows these circles (see attached picture).

Any ideas where this is coming from, if it is a bug or something I can avoid?


Pages: [1]