Hello michael96_27,

Yes, it appears rather "goofed up" in 14.9 on Windows.  It's the integration of the Arduino GUI which is causing the problem.  I looked into fixing it correctly, but got side tracked with the Kinect stuff.

Give me a few days and I'll make a new release.  Fortunately, I can test with Windows 7 64 bit now .. (although this particular issue spans more than just that version of the OS)

GroG

Just so you know, I've got to move all the options in the Arduino editor - here listed as "editor" to a non-Arduino editor named  at the moment "editor2". 

After this is done, it should be easy to enhance the new "editor2" with functionality like autocompletion...

Alrighty,
I just tested it on  Windows 7 64 bit... 
The repo is patched up..
And OpenCV & JavaCV have been updated.

You can download this "Bleeding Edge" Version by clicking the "All files in zip" on the build server's distribution page :

here -> http://myrobotlab.dyndns.org:8080/job/myrobotlab/ws/myrobotlab/dist/intermediate.729.20120820.0631/

• unzip it
• scroll down to OpenCV under the Runtime/Borg tab
• right-click - "install"
• wait a while - OpenCV has a lot of dependencies
• make sure it installs with a success
• scroll down to OpenCV under the Runtime/Borg tab - (again)
• right-click - "start" -  name your service, something creative like "camera"  or "opencv"
• go to the new tab
• make sure the input drop down is set to OpenCV - (this is the opencv frame grabber, good for most webcams)
• press "capture"
• if it works take a screen capture and post it !
• if it doesn't work send me the myrobotlab.log file from the same directory.

Going forward:
For things which don't work - please attach the myrobotlab.log file and try to provide as much information as possible. Answer my previous question : "So is your bot working now?  Is it a differential drive i.e. a motor on each wheel, or does it have some other configuration.  This is where pictures help.."

Here's your log snippet :

Notice you can successfully download .jar files but not .dll

I suspect that is it...
The dlls as well as all serice dependencies are on codegoogle.com -
I'm curious what happens if you try to manually download it ...

http://code.google.com/p/myrobotlab/source/browse/#svn%2Ftrunk%2Fmyrobotlab%2FthirdParty%2Frepo%2Fnet.sourceforge.opencv%2Fopencv%2F2.4.2%2Fx86.64.windows

click on one of the dlls .. do you get a popup or warning from the AVG software ?

You can't set format directly on the camera - there is a format parameter which can be set through Python - but some models & makes of cameras ignore it anyway.

You can resize the stream by using the PyramidDown PyramidUp to halve or double (respectively) the stream.

Alrighty.....
The "next step" - is to begin incorporating what you want into a script - I created this one so that you can get an idea of the possibilities.

You will need the latest bleeding edge build - since there was some filter stuff I had to fix. Additionally, there is probably a memory leak in FindContours, or the Jython console logging - I'll figure that out later, since my current focus is fixing up the Arduino service

from java.lang import String
from java.lang import Class
from org.myrobotlab.service import Runtime
from org.myrobotlab.service import OpenCV
from com.googlecode.javacv.cpp.opencv_core import CvPoint;
from org.myrobotlab.service import OpenCV

# create or get a handle to an OpenCV service
opencv = Runtime.createAndStart("opencv","OpenCV")

# add the desired filters
opencv.addFilter("PyramidDown1", "PyramidDown")
opencv.addFilter("InRange1", "InRange")
opencv.addFilter("Dilate1", "Dilate")
opencv.addFilter("FindContours1", "FindContours")

# change values of the InRange filter
opencv.setFilterCFG("InRange1","hueMin", "3")
opencv.setFilterCFG("InRange1","hueMax", "33")
opencv.setFilterCFG("InRange1","saturationMin", "87")
opencv.setFilterCFG("InRange1","saturationMax", "256")
opencv.setFilterCFG("InRange1","valueMin", "230")
opencv.setFilterCFG("InRange1","valueMax", "256")
opencv.setFilterCFG("InRange1","useHue", True)
opencv.setFilterCFG("InRange1","useSaturation", True)
opencv.setFilterCFG("InRange1","useValue", True)

# change value of the FindContours filter
# opencv.setFilterCFG("FindContours1","minArea", "150")
# opencv.setFilterCFG("FindContours1","maxArea", "-1")
# opencv.setFilterCFG("FindContours1","useMinArea", True)
# opencv.setFilterCFG("FindContours1","useMaxArea", false)

# ----------------------------------
# input
# ----------------------------------
# the "input" method is where Messages are sent to this Service
# from other Services. The data from these messages can
# be accessed on based on these rules:
# Details of a Message structure can be found here
# http://myrobotlab.org/doc/org/myrobotlab/framework/Message.html 
# When a message comes in - the input function will be called
# the name of the message will be msg_++_+
# In this particular case when the service named "opencv" finds a face it will publish
# a CvPoint.  The CvPoint can be access by msg_opencv_publish.data[0]
def input():
    #print 'found face at (x,y) ', msg_opencv_publish.data[0].x(), msg_opencv_publish.data[0].y()
    arrayOfPolygons = msg_opencv_publish.data[0]
    print arrayOfPolygons
    if (arrayOfPolygons.size() > 0):
   	print arrayOfPolygons.get(0).centeroid.x(),arrayOfPolygons.get(0).centeroid.y()
    return object

# create a message route from opencv to jython so we can see the coordinate locations
opencv.addListener("publish", jython.name, "input", CvPoint().getClass()); 

# set the input source to the first camera
opencv.capture()

What this script does :

• it creates and starts the an OpenCV service
• it adds the appropriate filters
• it configures the filters
• it sets a message route from the OpenCV service to the Jython service for "found polygons"
• for any found polygon, the Jython service prints out the first polygons center coordinates

Here is what happens on my laptop. 
You'll noticed the 4 filters are not displayed, this is because the service is not notifying the gui that new filters have been added, although, they were added programmatically.  At the moment, if you add a filter programmatically, you'll have to remove it programmatically.  Some gui notification details are low on my priority list.  So, whatever filter values you have assigned, I had something which fit the criteria,.Strangely, I think it was my hair :D 

It found 1 polygon and sent it back to the jython service.

So the polygon's center is printed out in the jython service's "jython console".
here you can see the x,y location of that polygon at (228,87)

Now, the Jython service could take those two values, adjust or modify them, and send them to an Arduino service which has 2 Servo's attached ....   and viola !  you have color tracking :)

Hope this makes sense.

BTW - this is getting very off topic, if you wouldn't mind could you create another forum entry if you have another question or preferrably create a page and post your Robot Project !  For an example look at DancesWithRobots' Houston entry - http://myrobotlab.org/content/houston

Disregard the 

"</sending></sending>" tags at the end of the script... This is being added by Drupal for some incorrect reason...