Re: Tracking Multiple Targets
 

The operations performed in image processing are expensive, and become even more so when allocating 300 kb (640x480 at 1 byte per pixel; a BinaryImage). Is it actually outputting processed data? If so, it may just be that the processing is taking too much time. There are a few ways to optimize the image processing, and you can do such things as setting up the processing to run in a separate thread and putting the drive code in continuous to make sure the robot drive gets updated often enough.

But before any of this, there's one thing I'd like to see: the output, in its entirety. If an exception happened, the type of exception and the call stack can provide a lot of information. If there's no exception, it just means you need to find some creative ways to optimize.

Re: Tracking Multiple Targets
 

I didn't copy the output, but I don't think there were any exceptions. What CRio did you guys use this year? We used the CRioI.

Re: Tracking Multiple Targets
 

We found that similiar code took about a half second to process an image - which is too long - the robot gets cranky. So we do the image processing in a seperate thread so the robot can keep processing other things while the image is being manipulated.