How should I calculate the depth accuracy I should expect?

[AndyO]

I have a robotic application where I want to be able to create G-Code to select and then describe each object on a 610mm x 457 region of focus. I can place 1 or more cameras within 1m and can optimize the height for the application. The ROF will be stationary at the time of image capture. I can control the lighting through curtains or adding additional lighting.

I need g-code that is accurate to the .25mm or better. Is this a reasonable expectation/target to have? What accuracy should I be able to reach reasonably?

How can I use the information on Intel Spec sheets to determine what I should expect from each camera?  Does running in stereo improve accuracy and if so, by how much?

Thanks,
Andy

[sentientSpark_Intel]

Hi AndyO,

Thank you for your post regarding your photogrammetry system! Are you by chance asking about the Intel® RealSense™ camera? If so, you will be more likely to get solutions to your questions by posting in this discussion board for Intel® RealSense™ in the link below: https://support.intelrealsense.com/hc/en-us/community/topics.

It looks like it is quite active, and I hope that you can get answers promptly.

This support forum was created to answer questions and help support collaboration regarding Intel Graphics Performance Analyzer (Intel's tool to improve a game's performance by quickly identifying problem areas), and if you have any questions about Intel GPA I would be happy to answer them!

Cheers!