Hi RobotMan,
There are many variables to take into consideration for finding and recognizing color, as well as multiple programming approaches. The tutorial introduces the basics of using the Vision Assistant to prototype a basic color location algorithm. This approach may not prove to be very robust, considering changing lighting conditions, motion blur, noise, and other variables. You may want to consider the color characteristics of the trailer identifiers, and how they can be differentiated from the background. Perhaps a color threshold and then binary morphology processing would work better for this application. It's hard to say what will work best right off the bat...it's really a problem that would benefit from trial and error.
If you are running a program from the cRIO to try to find the colors based on the templates generated in the examples, you will have to move the templates over to the cRIO as per this document:
FTP File Transfer to cRIO (Using Image Processing Requiring Template Images)
Also, the colors may not match sufficiently if you are searching for a different blue, for example. You would want to take a new template image of the exact object that you are trying to detect.
More information on color image processing in general can be found in the Vision Concepts Manual linked at the bottom of the Image Processing Tutorial that you referenced.
Hope this helps,
~Nate