Hello y’all!
I’ve been working on vision tracking for our robot as an off-season goal. I decided to it on the dashboard, so we could display the processed image as well as the original side by side. My code and some dashboard views are attached to this post.
When I go to actually track the images, if I choose not to process the image, and only show the original, everything (that should work) works fine. However, when I enable the vision, and start processing the image, both images go black (including the one for original image).
Once in a while, if I drag over the image, it flickers with the right image.
Is there a solution to the problem? I am especially concerned with the fact that the original image goes dark if I want to process, despite it not being connected with any wires downstream.
The image processing I remember doing on the DS laptop only has one actual image buffer. The processing steps replace the contents. If you’re doing a threshhold function to come up with a binary image for further analysis (find convex hull, look at aspect ratio, that sort of thing), your image will end up with values of 0 and 1, which are probably indistinguishably dark on your screen. To fix that, edit the properties of the display to use a binary palette. I think the “detected” pixels will then show up red by default.
If you want to see both pre- and post-processed images, you’re going to have to open a second IMAQ image buffer and tell the threshholding function to put its output in the second buffer instead of reusing the first one.
I will try that as soon as I am back in the workshop. Thank you!
