loading
 
Object tracking
Jim Ivy from United States  [8 posts]
12 year
Just purchased RoboRealm with AVM. I would like to uase RR and AVM to recognize movement and then track the moving object until it is no longer visible. Is this possible?

Jim
EDV  [328 posts] 12 year
I added “Learn from motion” option to “Object recognition” mode and you should download RoboRealm package with new AVM Navigator v0.7.4.

>> Hi, how do I update AVM Navigator? I purchased it last version.

You had to receive download link after registration on RoboRealm site.
Just use this link for downloading recent version of RoboRealm package with AVM Navigator.

You could also start RoboRealm application and then click "Option" button and further click "Download Upgrade".

Further you should set "Learn from motion" checkbox with "Object recognition" mode in AVM Navigator dialog window.
Jim Ivy from United States  [8 posts] 12 year
EDV,
Thanks I have not had time to really work with my project. But I have managed to get the movement & COG modules to work. I will be going on vacation soon and plan on working on my project then. Waiting for a PTZ camera at this time. Going for a weatherproof IR foscam.
EDV  [328 posts] 12 year
Here is complete solution of object tracking:
http://forums.trossenrobotics.com/showthread.php?4764-Using-of-AVM-plugin-in-RoboRealm&p=48865#post48865

I tested it on the prototype of “Twinky rover” and it works fine:
http://www.youtube.com/watch?v=pt2y7xkiTXo

See also “Twinky rover presentation”:
http://www.youtube.com/watch?v=Fkjpma1oWAg

Testing of new robot:
http://www.youtube.com/watch?v=F3u0rTNBCuA
EDV  [328 posts] 12 year
It is a testing of the enhanced tracking algorithm that takes into consideration variable servo speed:
http://www.youtube.com/watch?v=ueqDhuHiR-E

See VBScript program and diagram below for more details:


 

program.robo
EDV  [328 posts] 12 year
>> I've had a 'fiddle' with AVM navigator and managed to teach it a few objects/faces, but not the first idea how to turn that into tracking commands…

It is easy. You should just use "Object recognition mode" in AVM Navigator module.
First clear the AVM search tree by click on button "Set key image size(New)" and further press "Yes".

Now you can train AVM on some faces like in video below:
http://www.youtube.com/watch?v=oRpXBOX5rxY

When training will be done you should use variables that described below for your VBScript program:

NV_OBJECTS_TOTAL - total number of recognized objects
NV_ARR_OBJ_RECT_X - left-top corner X coordinate of recognized object
NV_ARR_OBJ_RECT_Y - left-top corner Y coordinate of recognized object
NV_ARR_OBJ_RECT_W - width of recognized object
NV_ARR_OBJ_RECT_H - height of recognized object

As example you can use these VBScript programs that was published in this topics:
http://www.roborealm.com/forum/index.php?thread_id=3881#
http://forums.trossenrobotics.com/showthread.php?4764-Using-of-AVM-plugin-in-RoboRealm&p=48865#post48865
EDV  [328 posts] 12 year
Bmw318be>> Hi, but how to do the localization in pan and tilt if I want to use pan rotation to do the localization in the training mode, am I be able to do automatically.

I think that this topic can help in your questions regarding object tracking.

Dawson from Malaysia  [173 posts] 12 year
Hi,

I am interested in AVM's "learn from motion" feature. Was it designed to work on a 2D plane or will it also work on a 3D plane i.e. object is moving closer or further away. I am currently using the blob tracking feature of RR to maintain a lock on the object being tracked. I am hoping that this AVM feature can help my project maintain a lock on the object in uneven lighting (outdoor) conditions.
EDV  [328 posts] 12 year
The good contrast of movable object on the general background and also size of motion area that commensurable with key image size (usually 80x80 pixels) are necessary conditions for working of "learn from motion" feature.

So, if these conditions will be provided in your case then perhaps it will be working.
Dawson from Malaysia  [173 posts] 12 year
Excelent
Dawson from Malaysia  [173 posts] 12 year
Impressive video
Anonymous 12 year
I'm doing artificial landmark tracking robot. There are four landmarks. i'm using shape match module. my robot is can recognized the landmark well while moving. It is possible when the robot see the landmark and capture the image from video, process the capture image then only continue moving.
if possible. How?

Thank you.
Anonymous 12 year
I'm not sure what you are asking. Do you want the robot to save the landmark image to disk before continuing? Or do you want it to process images from a video?

The Write_Images module may be of interest.

http://www.roborealm.com/help/Write_Images.php

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