Go Down

Topic: Facial Tracking... Infamous black screen (Read 1 time) previous topic - next topic


Heres what I am attemptiing to do. Take my eyes and get them to pan and track to someones face when they walk up or by.

These will be used in a halloween prop and while I think I am on the right track I wanted to reach out to the group to garner thoughts and opionins as to the process and if there is a better way by all means I am open.

Currently my plan is to use Windows 7 64bit OS: Webcam input to Processing + OpenCV talking to Arduino.

      Later I will attempt to:
Implement a IR Webcam
Transmit via WiFi or Ethernet
Expand to Audio too.
Last night I spent some time installing everything and ran into a problem (which I have read about) where I get a black screen from my webcam and a haarcascade file location issue. This is where I am stuck right now.

Thanks in advance!


I have gotten the haarcascade issue fixed but now am left with the black screen issue.

Tested on processing 2 64 and 32 bit and same result. I tried the v1.5 and it had library issues so wrote that off.

Screenshot included....

Here is the code:


Code: [Select]
import processing.video.*;
import hypermedia.video.*; //import the necessary libraries
import java.awt.Rectangle;
import processing.serial.*;
import cc.arduino.*;

Capture myCapture;
OpenCV opencv;   //create the library object

// contrast/brightness values
int contrast_value    = 0;
int brightness_value  = 0;

// The arduino object:
Arduino arduino;   



void setup() {

    size( 320, 240 ); //set window size
    //windows fix addition   
    myCapture = myCapture = new Capture(this, width, height, 30);

    opencv = new OpenCV( this ); //setup openCV
    //windows fix addition
    opencv.allocate(width, height);
    //opencv.capture( width, height );   // open video stream
    //the arduino serial setup
    arduino = new Arduino(this, Arduino.list()[0], 57600);
    //list of available detection types: http://ubaa.net/shared/processing/opencv/opencv_cascade.html
    opencv.cascade( OpenCV.CASCADE_FRONTALFACE_ALT );
    //opencv.cascade( OpenCV.CASCADE_PROFILEFACE);
    //opencv.cascade( OpenCV.CASCADE_FULLBODY);

    // print usage
    println( "Drag mouse on X-axis inside this sketch window to change contrast" );
    println( "Drag mouse on Y-axis inside this sketch window to change brightness" );




void draw() {

    //windows fix addition
    opencv.read(); //read a new camera frame
    opencv.convert( GRAY ); //convert frame to gray scale
    opencv.contrast( contrast_value ); //apply contrast filter
    opencv.brightness( brightness_value ); //apply brightness filter

    // detect faces and put them in an array
    Rectangle[] faces = opencv.detect( 1.2, 2, OpenCV.HAAR_DO_CANNY_PRUNING, 40, 40 );

    // display gray image
    image( opencv.image(), 0, 0 );

    // draw face detection rectangles
    if(faces.length > 0)
      //the RC servo only follows the first face found.
      int mappedValue = (int) map(faces[0].x,0,width, 0, 255); //map from screen width to 255
      arduino.analogWrite(3, mappedValue); //send data to the arduino
    for( int i=0; i<faces.length; i++ ) {
        rect( faces[i].x, faces[i].y, faces[i].width, faces[i].height );             


  change brightness and contrast values

void mouseDragged() {
    contrast_value   = (int) map( mouseX, 0, width, -128, 128 );
    brightness_value = (int) map( mouseY, 0, width, -128, 128 );



public void stop() {
    opencv.stop(); //shutdown opencv on exit

And the Arduino code is just the firmata:

Code: [Select]
#include <Servo.h>
#include <Firmata.h>

Servo servos[MAX_SERVOS];

void analogWriteCallback(byte pin, int value)
    if (IS_PIN_SERVO(pin)) {

void setup()
    byte pin;

    Firmata.setFirmwareVersion(0, 2);
    Firmata.attach(ANALOG_MESSAGE, analogWriteCallback);

    for (pin=0; pin < TOTAL_PINS; pin++) {
        if (IS_PIN_SERVO(pin)) {

void loop()

Any ideas?


i need the hypermedia do you have a link where i can get this ? i have some examples i found on youtube which i cant use cuz of the lack of this import


i can help with what your trying to do since im doing something similar but with 2 webcams for depth perceptions by triangulating from a compare of two images

Go Up

Please enter a valid email to subscribe

Confirm your email address

We need to confirm your email address.
To complete the subscription, please click the link in the email we just sent you.

Thank you for subscribing!

via Egeo 16
Torino, 10131