This code was developed for the Preston HackSpace 2013 BarCamp stall in Blackpool
To run this code, you need the following installed
Python2.7 PySerial PyFirmata
Instructions Install PyFirmata Standard to an ArduinoMega
Attach the Arduino to the computer Connect a Servo to Pin 7 (x axis - left and right) Connect a Servo to pin 6 (y axis - up and down)
Install a WebCam and attach it to the servo horns (In what ever way you can, but make sure it is stable)
Run track.py - This will start the camera which will by default be looking for a green color. Run firmata.py - This will start the connection to the Arduino.
More Detail.
track.py writes to the values.txt file the coords of the color it is tracking in relation to the webcam window. firmata.py reads the values.txt file and trys to get the object into the center of the screen by moving the servo with the camera attached.
To Change the color alter the folloing line in track.py
cv.InRangeS(hsv_img, (38, 80, 80), (75, 255, 255), thresholded_img)
Alter the first number in each tuple based on the list below. Orange 0-22 Yellow 22- 38 Green 38-75 Blue 75-130 Violet 130-160 Red 160-179
For more information on this google "openCV hue range for the HSV color model"
For more information on this project you can vist it the projects home page http://wiki.prestonhackspace.org.uk/index.php?title=Face_and_Color_tracking_web_cam
Have Fun!!!!
Update 18th June 2013: Fixed: bug where was not centering on the screen Added: The face tracking pick the face nearest the center of the screen and ignores the rest Added: Colours all faces Red for icgnored and Green as focusted