The program is really simple. I took a string of words, split them up using char and took those individual characters/letter and put them in an array. Then I made an array for the y position of each letter. The program just checks the y position of each letter as it falls and overlays that information with the pixel array of the video. If the y position of the letter is on a pixel in the video that is darker than a set threshold, then the speed of the letter reverses (which basically results in the letters seemingly staying in place when in between a dark and a light pixel, or between the background and an object).

Github Repository: https://github.com/yvonnehidle/textrain
Original Blog Post @ Arealess: http://www.arealess.com/text-rain-version-1/

//////////////////////////////////////////////////////////////////
// GLOBAL VARIABLES
//////////////////////////////////////////////////////////////////
import processing.video.*;
Capture video;

// string array for poem
String poem = "Fancy lines and dancing swirls have nothing on simplicitys curls";
char[]letters = new char[poem.length()];

// letter positioning
final int max = letters.length;
float[]letterY = new float[max];
//////////////////////////////////////////////////////////////////

//////////////////////////////////////////////////////////////////
// BASIC SETUP
//////////////////////////////////////////////////////////////////
void setup()
{
// general setup
size(640,480);
noStroke();
textSize(20);
fill(255,0,0);

// video
video = new Capture(this,640,480,30);

// divide poem into individual characters
for(int i=0; i<poem.length(); i++)
{
letters[i] = poem.charAt(i);
}
}
//////////////////////////////////////////////////////////////////

//////////////////////////////////////////////////////////////////
// TABLE OF CONTENTS
//////////////////////////////////////////////////////////////////
void draw()
{
background(255);

// falling letters
fallingLetters();
}
//////////////////////////////////////////////////////////////////

//////////////////////////////////////////////////////////////////
// FALLING LETTERS
//////////////////////////////////////////////////////////////////
void fallingLetters()
{
// if a webcam is available, load the video
if(video.available())
{
video.read();
}
video.filter(GRAY);
image(video, 0, 0);

// variables
float letterS=1;
float letterX=0;
float letterXSpace=width/letters.length;
int darknessThreshold = 180;

// generate the letters and have them interact
// pixel by pixel of the video
video.loadPixels();

for(int i=0; i<letters.length; i++) {

// what is the pixel number in the array?
int loc = int( letterX + letterY[i] * video.width );

// draw letters
text(letters[i],letterX,letterY[i]); letterX=letterX+letterXSpace;

// if the letters reach the bottom of the screen, start them at top again
if(letterY[i] >= video.height - 1)
{
letterY[i] = 0;
}

// if the brightness of the pixel is less than our darkness threshold
// then do not move the letter
else if(brightness(video.pixels[loc]) < darknessThreshold) { if(letterY[i] > 10)
{
letterY[i]-=letterS;
}
}

// else always move the letter
else
{
letterY[i]+=letterS;
}
}

}
//////////////////////////////////////////////////////////////////