Forum

Author Topic: Calculate estimated Exterior Orientation parameters  (Read 1631 times)

Hrishi

  • Newbie
  • *
  • Posts: 2
    • View Profile
Calculate estimated Exterior Orientation parameters
« on: January 20, 2021, 10:36:47 AM »
Hi,

I am trying to understand the python script provided here:
https://agisoft.freshdesk.com/support/solutions/articles/31000145016-how-to-calculate-estimated-exterior-orientation-parameters-for-the-cameras-using-python

I have two question related to the script:

1. It's been mentioned that matrix m is "transformation matrix to the LSE coordinates in the given point"; what does LSE stand for and what does matrix m signify?

2. Can someone help me understand the significance of the multiplying diagonal matrix Metashape.Matrix().Diag([1, -1, -1, 1])
to T * camera.transform? I understand what T * camera.transform signifies, but don't understand the multiplication with diagonal matrix

Thank you!

Alexey Pasumansky

  • Agisoft Technical Support
  • Hero Member
  • *****
  • Posts: 14968
    • View Profile
Re: Calculate estimated Exterior Orientation parameters
« Reply #1 on: January 20, 2021, 05:03:48 PM »
Hello Hrishi,

LSE is Local Space Euclidean coordinate system. Its orthogonal (Cartesian) system which doesn't consider any corrections related to the Earth curvature and etc.

Multiplication by (1, -1, -1, 1) diagonal matrix is required to switch from Metashape coordinate system orientation convention to another one required to estimate exterior orientation parameters in geographic/projected coordinate system.
In classical camera coordinate system rotation angles are defined around the following axes:
X axis is directed to the right of the image,
Y axis is directed to the top
and Z axis - backward. Whereas in Metashape in camera coordinate system Z is pointing to the camera view direction.
Best regards,
Alexey Pasumansky,
Agisoft LLC