First Test OpenCV with MRL services

My first test successfully ... all with the help of the developers of the MRL. 
As you can see .. More than 10 000 frames .. all that to properly configure the Pan / Tilt hahah
I can only thank .. and let's work!
http://www.youtube.com/watch?v=gRv0nmyECns&feature=youtu.be
 
 
 
 
 
 
 

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
GroG's picture

WOOOHOOO ! High Five o/

WOOOHOOO !

High Five o/

Saymon Cristian Alves Oliveira's picture

\o hahahah #More Stepsssss

yes

Saymon Cristian Alves Oliveira's picture

\o hahahah #More Stepsssss

yes

Alessandruino's picture

Nice :)who hooo :) which

Nice :)who hooo :) which script you used?

Saymon Cristian Alves Oliveira's picture

The basic Script...

The script I used is in the video description .. 
It was ready in the DSU tutorials I changed only a few things ... (Filters) 
Below the script:

This is the script I used to work, past in Python service:



port = "COM3"

xServoPin = 7

yServoPin = 6



tracker = Runtime.createAndStart("tracker", "Tracking")



# set specifics on each Servo

servoX = tracker.getX()

servoX.setPin(xServoPin)

servoX.setMinMax(30, 150)



servoY = tracker.getY()

servoY.setPin(yServoPin)

servoY.setMinMax(48, 180)



# optional filter settings

opencv = tracker.getOpenCV()



# setting camera index to 1 default is 0

opencv.setCameraIndex(1) 



# connect to the Arduino

tracker.connect(port)



# Gray & PyramidDown make face tracking

# faster - if you dont like these filters - you

# may remove them before you select a tracking type with

# the following command

# tracker.clearPreFilters()



# diffrent types of tracking



# simple face detection and tracking

# tracker.faceDetect()



# lkpoint - click in video stream with 

# mouse and it should track

tracker.startLKTracking()



# scans for faces - tracks if found

# tracker.findFace()

kmcgerald's picture

Great Video!

I like the picture in picture effect.  The funny thing was that I was watching the video of the ball and thinking, "nothing is happening" and then I glanced back over to the pan-tilt camera and saw it was moving around.  Duh!  If it's working correctly, the ball should look like it's not moving.  I guess if the wall had more features on it I wouldn't have been tricked.

It was interesting that the camera followed the ball even after it saw your face.  I'm guessing the tracking was tied to the lkoptical filter and not the face detect filter.

Great job. Keep up the good work. You've done more with MRL than I have so far.

Saymon Cristian Alves Oliveira's picture

hohoho Thanks!!

It was fun to edit video after it's done ... hahah makes you think nothing is happening since the camera is stopped and the orange ball is in another video .. does are two channels of videos made in same time.
 
I thought about writing on the wall with nothing lest external factors ( dirt , objects ) intrude recognition .
The filters that I used was LKOptical along with FaceDetect ...
I'm making progress thanks to all of you .. and always want to improve and help everyone here MRL ...
Thanks for the comment ! And let's work ! laugh