Forum

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 - karad

Pages: 1 [2] 3
16
General / fiducial marks?
« on: September 25, 2017, 08:46:55 AM »
hi,
I typed in  the coordinates of fiducial marks and the focal length but   No message to measure the fid.  marks on the photos....

17
I think your model exibits the well known "dome effect" (see literature), caused by insufficient calibration of the camera..

18
General / Re: Errors of calibrated parameters
« on: July 24, 2017, 12:09:59 PM »
 :)

20
General / Re: Errors of calibrated parameters
« on: July 21, 2017, 01:50:37 PM »
Hi Courtois!

Very interesting results!
Which is your GCP configuration and check points position?
Because 8.8 mm FL is too short, eg super wide angle, you have to use another lens distortion model as suggested by photoscan?

Regards

23
General / Re: Errors of calibrated parameters
« on: July 17, 2017, 11:23:23 PM »
hi,
of course we don not know at the beginning of the block adjustment the amount of  correlation. After the first run you remove the parameter putting 0 as you say. I think this must be the way Agisoft removes parameters.
regards

24
General / Re: Errors of calibrated parameters
« on: July 17, 2017, 01:05:39 PM »
Hi Courtois,
high correlation means doublicating the unknown parameter e.g. high correlation between K1 and K2 means that K1 is identical to K2 and must be removed in order to avoid instability of the solution. In this case the physical  interpretation is that the radial distortion is modeled  using only the K1 parameter for the particular camera.
The issue of correlated additional parameters is an old one in aerial bundle block adjustment and and as far as I know is addressed  removing high correlated parameters.
Regards

25
General / Re: Errors of calibrated parameters
« on: July 17, 2017, 11:08:31 AM »
Hi,
I think you sould consider the high correlation 0.99 ( as in COURTOIS 'S EXAMPLE)  between ADDITIONAL parameters. High correlation means overdetermination that causes ill conditioning of the solution of the normal equation system and unstable results. You must remove one of the highly correlated parameters.

26
Usefull :
1.  Nasrullah, A.R. (2016) Systematic analysis of unmanned aerial vehicle (UAV) derived product quality. Enschede, University of Twente Faculty of Geo-Information and Earth Observation (ITC)
2. Photogrammetric Computer Vision: Statistics, Geometry, Orientation and Reconstruction  2016,  Wolfgang Förstner (Author), Bernhard P. Wrobel (Author)

27
General / Re: "Not enough memory"
« on: May 03, 2017, 08:51:56 AM »
I had a project with 232 photos  20 mb each , settings arbitrary - high quality using a pc i7 5930k 64 gb ram -1070 8 gb and finished in about one hour. IN YOUR CASE I THINK THE PROBLEM ARISES probably due to the leafs of the trees. THEY GENERATE TOO MANY POINTS that can not be correlated and the matrices produced are huge and require much memory. Mask the areas with trees and see what happens.












28
General / Re: Can't build my first model
« on: March 18, 2017, 11:08:59 PM »
Hi,

check overlap first, forward 80%, sidelap 60%. On corners at least 3 photos at 10 degrees separation of taking position . If this is ok, try to check if  photos are stored in good sequence, else change the filenames to fix.

Dimitri

29
General / Re: Camera recommendations
« on: March 18, 2017, 10:50:48 PM »
Sony alpha 5000 around 360 euros 16-50 mm lens included

30
General / INCONSISTENCY....
« on: March 07, 2017, 10:08:34 PM »
 ?

Pages: 1 [2] 3