Aligning particles to a processed image

In this recipe, we will show how you can use techniques you were introduced to in the previous recipes to make particles align to the edge detected in the image.

Getting ready

In this recipe, we are going to use the particles' implementation from the Creating a particle system in 2D recipe; the image processing example from the Detecting edges recipe in Chapter 3, Using Image Processing Techniques; as well as simulating repulsion covered in the Applying repulsion and attraction forces recipe.

How to do it…

We will create particles aligning to the detected edges in the image. Perform the following steps to do so:

  1. Add an anchor property to the Particle class in the Particle.h file.
    ci::Vec2f anchor;
  2. Set the anchor value at the end of the Particle class constructor in the Particle.cpp source file.
    anchor = position;
  3. Add a new property to your application's main class.
    float maxAlignSpeed;
  4. At the end of the setup method, after image processing, add new particles, as follows:
    mMouseDown = false;
repulsionFactor = -1.f;
maxAlignSpeed = 10.f;

mImage = loadImage( loadAsset("image.png") );
mImageOutput = Surface8u(mImage.getWidth(), mImage.getHeight(), false);

ip::grayscale(mImage, &mImage);
ip::edgeDetectSobel(mImage, &mImageOutput);

Surface8u::Iter pixelIter = mImageOutput.getIter(Area(1,1,mImageOutput.getWidth()-1,mImageOutput.getHeight()-1));

while( pixelIter.line() ) {
    while( pixelIter.pixel() ) {
        if(pixelIter.getPos().x < mImageOutput.getWidth()
          && pixelIter.getPos().y < 
          mImageOutput.getHeight()
          && pixelIter.r() > 99) {
            float radius = 1.5f;
            float mass = Rand::randFloat(10.f, 20.f);
            float drag = 0.9f;
            Particle *particle = new Particle( 
            pixelIter.getPos(), radius, mass, drag );
            mParticleSystem.addParticle( particle );
        }
    }
}
  5. Implement the update method for your main class, as follows:
    void MainApp::update() {
  for( std::vector<Particle*>::iterator it = mParticleSystem.particles.begin(); it != mParticleSystem.particles.end(); ++it ) {

    if(mMouseDown) {
      Vec2f repulsionForce = (*it)->position - getMousePos();
      repulsionForce = repulsionForce.normalized() * math<float>::max(0.f, 100.f - repulsionForce.length());
                  (*it)->forces += repulsionForce;
        }

    Vec2f alignForce = (*it)->anchor - (*it)->position;
    alignForce.limit(maxAlignSpeed);
        (*it)->forces += alignForce;
    }

  mParticleSystem.update();
}
  6. Change the draw method for Particle inside the Particle.cpp source file to read as follows
    void Particle::draw(){
  glBegin(GL_POINTS);
  glVertex2f(position);
  glEnd();
}

How it works…

The first major step was to allocate particles at some characteristic points of the image. To do so, we detected the edges, which was covered in the Detecting edges recipe in Chapter 3, Using Image Processing Techniques. In step 4 you can see that we iterated through each pixel of each processed image and placed particles only at detected features.

You can find another important calculation in step 5, where we tried to move back the particles to their original positions stored in the anchor property. To disorder particles, we used the same repulsion code that we used in the Applying repulsion and attraction forces recipe.

How it works…

See also

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
18.226.34.25