Problems with GRIP

Hello everybody,
I’m trying to implement vision processing via GRIP and usb camera.
I’m geeting confusing to get the shape of the tower’s lights.–> |_|
When I do HSL, HSV, RGB and more filters to find this shape but its find some more crap.
There is a web who teach all this GRIP software, and all of it’s options, how to find custom shaps, filder between different shapes and more? - a web/place who shows how to do all of this?


On the wpilib documentation, there are some instructions for using GRIP with this year’s game: