none
What range of surface temperatures does the Kinect return accurate depth readings? RRS feed

  • Question

  • I am involved in a project at the University of Technology Sydney and will be using the Kinect to measure distance information from the Kinect to a surface. The surface, being metallic, is quite likely to get hot, with temperatures occasionally exceeding 60 degrees Celsius. I am wondering whether the reliability of readings in such conditions has already been measured and if so, what the results were.

    Cheers.

    Wednesday, June 22, 2011 2:38 AM

Answers

  • Highly doubt you will have any kind of interference at 60ºC, because, at that temperature, the black body radiation intensity in the wavelenght used by kinect laser (around 830nm) should be still very low compared to the dots emited by the IR laser.

    Thursday, June 30, 2011 10:32 AM

All replies

  • pdquin, do you mean that (a) the surface where the kinect is standing will be approx 60C, or that (b) the temperature of the surface away from kinect, for which depth is being measured, will be 60C? I think you mean (b), but I'm just double checking.

    If you mean (a), then it seems a little risky, because the kinect camera will completely turn off if temperature reaches 90C, and if temperatures occasionally exceed 60C, then that plus other environmental factors may get kinect device temperature dangerously close to 90C.

    If you do mean (b), we haven't specifically tested how characteristics of depth perception change with temperature of surface being perceived but, as long as device is far enough away from hot surface to not cause a hardware malfunction, we don't expect this to be an issue.

    Hope this helps,
    Eddy


    I'm here to help
    Thursday, June 23, 2011 9:53 PM
  • Thanks Eddy,

    I do mean (b), though I suspect that the ambient temperature (i.e. the air around the Kinect) will get quite hot, though how much of that heat will then bleed into the Kinect, I don't know.

    Just for clarification, my concern is that the IR sensor used for depth readings will be affected by the massive emissions of IR from the metallic surface (since it will be hot and therefore emit more IR than at lower temperatures). Unless I'm misremembering my physics (which I very well could be).

    Cheers,

    Phillip



    Thursday, June 30, 2011 12:49 AM
  • Thanks Eddy,

    I do mean (b), though I suspect that the ambient temperature (i.e. the air around the Kinect) will get quite hot, though how much of that heat will then bleed into the Kinect, I don't know.

    Just for clarification, my concern is that the IR sensor used for depth readings will be affected by the massive emissions of IR from the metallic surface (since it will be hot and therefore emit more IR than at lower temperatures). Unless I'm misremembering my physics (which I very well could be).

    Cheers,

    Phillip



    what if you put a "glass wall" between the kinect and this surface? maybe the heat will not affect the kinect because you can somehow refrigerate the air around kinect....

    but I don't know if the glass wall will interfere with the kinect's IR

    Thursday, June 30, 2011 8:42 AM
  • Highly doubt you will have any kind of interference at 60ºC, because, at that temperature, the black body radiation intensity in the wavelenght used by kinect laser (around 830nm) should be still very low compared to the dots emited by the IR laser.

    Thursday, June 30, 2011 10:32 AM
  • Thanks. Given that it's hard to find specs for what wavelength IR the Kinect is emitting, I was finding it hard to confirm beyond reasonable doubt that the Kinect will be unaffected.
    Tuesday, July 12, 2011 3:52 AM