We’ve always run vision-processing on the cRIO (we code in Java). This time around, because of many different resources available to us, we would like to incorporate a laptop into this. Using a laptop would take quite a bit of work from the cRIO, and would make things smoother for us.
What we would like to do is to take information from our camera to a laptop (on the robot or off, it doesn’t matter. We can do this already). Then, after processing the image received from the camera, we would send, in the least, a processed image to the cRIO, and at most, some simple data to the cRIO in order to help determine relation to the backboard and hoops.
We know that network tables exist, but are unsure how to use them, especially if we use LabVIEW, which may provide us with a larger library for vision processing than Java.
Basically, how would we go about sending data from a laptop to the cRIO during teleop or auton?
I found the following document: http://firstforge.wpi.edu/sf/docman/do/downloadDocument/projects.wpilib/docman.root.c_and_java_documentation/doc1297/3
On page 85, it discusses “Laptop based vision processing.” However, it is poorly documented, with blank pages where there should be examples for finding circular targets via laptop-based machine vision.
I did so:
(Post 5)