Hi! I am using this code from this topic: http://forum.processing.org/topic/poor-opencv-camera-resolution

1. import hypermedia.video.*;
   import codeanticode.gsvideo.*;
   import java.awt.*;
   
   
   OpenCV opencv;
   GSCapture cam;
   
   int w = 800;
   int h = 600;
   int framerate = 60;
   int threshold = 80;
   
   boolean find=true;
   
   PFont font;
   
   void setup() {
   
       size( w*2+30, h*2+30 );
      
      
      
     
   
       opencv = new OpenCV( this );
       //opencv.capture(w,h);
      
       //My cam should be listed - otherwise GSVideo lib probably not working
       String[] cameras = GSCapture.list();
    
       if (cameras.length == 0)
       {
           println("There are no cameras available for capture.");
           exit();
         } else {
           println("Available cameras:");
           for (int i = 0; i < cameras.length; i++)
             println(cameras[i]);
           cam = new GSCapture(this, w, h, cameras[0]);
           cam.frameRate(framerate);
         }
         //Get some space for our own Image
         opencv.allocate(w,h);
        
         font = loadFont( "AndaleMono.vlw" );
         textFont( font );
    
         println( "Drag mouse inside sketch window to change threshold" );
         println( "Press space bar to record background image" );
    
     }
    
    
    
     void draw() {
       if (cam.available() == true) {
        
           background(0);
           //don't read - load from other cam
           //opencv.read();
           cam.read();
           opencv.copy(cam);
          
           //opencv.flip( OpenCV.FLIP_HORIZONTAL );
      
           image( opencv.image(), 10, 10 );                // RGB image
           image( opencv.image(OpenCV.GRAY), 20+w, 10 );   // GRAY image
           image( opencv.image(OpenCV.MEMORY), 10, 20+h ); // image in memory
      
           opencv.absDiff();
           opencv.threshold(threshold);
           image( opencv.image(OpenCV.GRAY), 20+w, 20+h ); // absolute difference image
      
      
           // working with blobs
           Blob[] blobs = opencv.blobs( 100, w*h/3, 20, true );
      
           noFill();
      
           pushMatrix();
           translate(20+w,20+h);
          
           for( int i=0; i<blobs.length; i++ ) {
      
               Rectangle bounding_rect    = blobs[i].rectangle;
               float area = blobs[i].area;
               float circumference = blobs[i].length;
               Point centroid = blobs[i].centroid;
               Point[] points = blobs[i].points;
      
               // rectangle
               noFill();
               stroke( blobs[i].isHole ? 128 : 64 );
               rect( bounding_rect.x, bounding_rect.y, bounding_rect.width, bounding_rect.height );
      
      
               // centroid
               stroke(0,0,255);
               line( centroid.x-5, centroid.y, centroid.x+5, centroid.y );
               line( centroid.x, centroid.y-5, centroid.x, centroid.y+5 );
               noStroke();
               fill(0,0,255);
               text( area,centroid.x+5, centroid.y+5 );
      
      
               fill(255,0,255,64);
               stroke(255,0,255);
               if ( points.length>0 ) {
                   beginShape();
                   for( int j=0; j<points.length; j++ ) {
                       vertex( points[j].x, points[j].y );
                   }
                   endShape(CLOSE);
               }
      
               noStroke();
               fill(255,0,255);
               text( circumference, centroid.x+5, centroid.y+15 );
      
           }
           popMatrix();
        
     }
   
   }
   
   void keyPressed() {
       if ( key==' ' ) opencv.remember();
   }
   
   void mouseDragged() {
       threshold = int( map(mouseX,0,width,0,255) );
   }
   
   public void stop() {
       opencv.stop();
       super.stop();

But I have some problems..

Until I use "Blob[] blobs = opencv.blobs( 100, w*h/3, 20, true );" all is going ok. opencv.image(); is colour RGB, and remember function works correctly.

BUT, when I including this line "Blob[] blobs = opencv.blobs( 100, w*h/3, 20, true );", code is going crazy!

0) actually, it's because of opencv.blobs(); function exactly... I can't find solution on forums..
1) opencv.image() gimme grayscaled image
2) remember() function doesnot work anymore...

when I use opencv.remember(); or opencv.remember(OpenCV.SOURCE); , memory image is black! only one pixel in first line of image is pink (hmm..)

if I use opencv.remember(OpenCV.BUFFER);  - I get some strange picture, white pixels on the edges of contrast shapes...

So, I can't use absDiff(); function correctly too because of this bug...
Can you help me?

p.s. I use: windows XP SP3, processing 1.2.1, opencv 1.0, gsvideo 0.4.5-win. As I know, opencv >1.0 doesn't support processing, am I right?

p.p.s. Usage of opencv.capture instead of GSCapture gimme workable functions, RGB image, but  verrry slow framerate...

Can you test this code on your computer, maybe it's depends on windows version...