loading
 
Changing Modules in the Stream
Anonymous from United States  [28 posts]
12 year
Is there any way from the VBScript module, or another module, to enable/disable modules running in the video stream?

I have a couple of different ways I need to detect objects, a color filter may work on some, while line detection may work on others, and they are all different colors. (Will be attempting the object detection module soon as well), but incase I ever need to use a bunch of modules, and only enable certain ones depending on the items I'm looking for, is there a way in code to do this?

I would assume something like Disable(Module) or Enable(Module), since there is a disable button. May could even use the VBScript to press the disable button, but how do you know which module is currently highlighted?

Thanks for any response!
Anonymous 12 year
Chris,

As apposed to thinking of disabling or activating modules think of it as executing it or not. The If statement module can be used to do this. Based on a variable of your chosing you can allow modules within the if statement module start and stop to run or not. So your VBScript can then set variables to allow specific modules to run or not. The disable/enable are really meant for users to use. Its easier to use conditional variables to execute or skip certain modules. See the attached robofile for an example:

STeven.
program.robo
Anonymous from United States  [28 posts] 12 year
Steven,
   That's great! I completely forgot about the IF statements for the pipe.

Now, if I may pick your brain a bit more, I'm trying to get the object detection module to work correctly. Is it better to have training pictures with a black background, or white background, or some random crazy color, or transparent?

I have to identify quite a few objects, and I'm really new to all of the different features. I would be able to talk privately about the code with you, as this is a very private project, but something to get me on the right track would be awesome.

I have trained the object recognition module with a few different pictures, and it will detect the objects about 10% of the time, but I have to have them in a perfect spot before it will lock on to them. If I'm too close to the camera, or too far away, it doesn't work correctly.

Thanks!
Anonymous 12 year
Chris,

The ideal background is one without any texture pattern. So a black or white or gray is fine ... just not one with any texture.

If you'd like you can send us a quick email via the Contact form so we can get your email address and continue offline to get some images from you to understand better why the recognition is so poor.

http://www.roborealm.com/contact.php

STeven.
Anonymous from United States  [28 posts] 12 year
Email sent! Thanks Steven!

This forum thread has been closed due to inactivity (more than 4 months) or number of replies (more than 50 messages). Please start a New Post and enter a new forum thread with the appropriate title.

 New Post   Forum Index