Hi guys, i am trying to process aero-triangulation on 10,000 (9,251) Drone images of 20 megapixel each . 186 gigapixel.
Out of 9251 images, about 3500 have not aligned. About 9 GCPs are used , just to place the model in correct position approximately.
3D grid: camera angle 45 degree with 70-70 overlap.
Workstation specifications: Intel(R) Xeon(R) W-2133 CPU @ 3.60GHz
128GB RAM
Nvidia Quadro P2000
Below is attached 2 pictures:
1. Total imported images
2. Aligned and failed images -- Aligned in Green. Failed in Orange
NOTE: Only Blue GCP's (9) are marked and used for processing. When processing Aero-triangulation GCP method was selected.
Any input or help is greatly appreciated!
Thank you!
For such large scale projects good 3D precalibration of cameras is mandatory, especially if nadir only photos are used - https://communities.bentley.com/products/3d_imaging_and_point_cloud_software/w/wiki/28536/the-model-appears-to-be-curved-instead-of-being-horizontal-after-at
Pleas note that GCPs in Contextcapture do not improve camera calibration(not sure why) only tries to warp the model on top of GCPs so by doing good precalibration GCPs can be used only as checkpoints to verify that calibration is accurate.