Hey there,

I ordered an OAK-D-Pro PoE and I am trying to detect thin structures (something like straws - range 4 - 8mm). Based on the depth information I create pointclouds to analyze these spatial structures.

The problem is, that the quality of depth information is not good enough for my purpose. So is there any way to improve it a little further? I've already tried setExpandedDisparity(True), better illumination and wls-filtering. Everything helps a little but doesn't seem to be the right solution.

Is there a chance to improve depth quality, even with cost of runtime? I appreciate every help.

  • erik replied to this.

    Thanks for your help @erik.

    I'll go through the tutorial and give you some feedback 😉

    2 months later

    So after all the time I'm back with my feedback.

    It didn't solve the accuracy in a way I wanted it to be, but it was getting a little bit better by changing the kernel of the census-trnasformation with: depth.initialConfig.CensusTransform.KernelSize.KERNEL_7x9

    But I think I wont get the accuracy that I need by stereovision, because census transformation (or disparity calculation) isn't accurate enough, even if the images are very well detailed.

    Nevertheless thanks for your help.