Getting different point cloud coordinates using the API and SOPAS

[magarcia-deuser]

Hello,

I am using a Visionary T-Mini camera and after running the continuous streaming samples in both C++ and Python, I've noticed that the point cloud they save is not the same as the one generated by SOPAS, yielding different X, Y and Z coordinates for the same point, with a few centimeters of difference in some cases. If I measure the real distances from the center of the sensor, SOPAS seems to be pretty accurate. Do I need to do any extra steps using the API to get the same point cloud as SOPAS? I am planning to use the camera in a robotics application where getting accurate measurements is crucial.

Thank you for your help.

[zimmefaSICKAG]

Hello,
we also received your question via your local SICK contact, which we will use as the main communication channel.
The final conclusion will also be posted here to eventually help others.

[yanchlu]

I met the same issue, how did they solve this problem?

[zimmefaSICKAG]

In the specific case mentioned here, we couldn't find any actual differences between the point clouds. However, the scaling needed to be adjusted as one point cloud was in units of mm while the other was m. Also a rotation of 180° was necessary to fully align both point clouds. We are currently investigating why this was necessary.