Hi again,
Thank you Yoann, but I need to avoid this image pairs before building the sparse point cloud.
I have been testing the matchPhotos() function and the results are the same when I use the parameter pairs() and when I don't use it. I explain:
To get the image pairs tuple list I do the matching process in a common project and I save the list of image pairs created. Then I edit the list removing the image pairs which I don't want to create. Finally I restart the entire project using my edited list in the matching process and I get image pairs which I don't want to create AGAIN. Why does it happen?
The edited list which I use in the pairs() parameter is like this:
[(<Camera 'IMG_170606_100743_0218_REG.TIF'>, <Camera 'IMG_170606_100748_0221_NIR.TIF'>),
(<Camera 'IMG_170606_100743_0218_REG.TIF'>, <Camera 'IMG_170606_100745_0219_REG.TIF'>),
(<Camera 'IMG_170606_100743_0218_REG.TIF'>, <Camera 'IMG_170606_100746_0220_REG.TIF'>),
(<Camera 'IMG_170606_100743_0218_REG.TIF'>, <Camera 'IMG_170606_100745_0219_GRE.TIF'>),...]
And the used code:
chu.matchPhotos(accuracy=PhotoScan.HighestAccuracy, preselection=PhotoScan.Preselection.NoPreselection, generic_preselection=False, reference_preselection=False, keypoint_limit=40000, tiepoint_limit=4000, pairs=matchesCam1List)
I have tried with True for generic and reference preselection too, and not writing the parameter "preselection" because I supposed it is obselete truly (The API document say:
• preselection (PhotoScan.Preselection) – Image pair preselection method (obsolete)), but it dosen't work for anything.
Am I doing it correctly?
Any idea?
Thank's anyway
