P5.js : manually calculate additive blendmode with pixels[] - processing

I’m currently coding a P5.js program in which I draw some random generated patterns like this :
function setup() {
colorMode(RGB, 255, 255, 255, 255);
blendMode(ADD);
}
function draw() {
for(var i = 0; i < 20000; i++)
{
x = generateCoordinatesX();
y = generateCoordinateY();
var pointColor = generateColor(); //rgba
stroke(pointColor);
point(x, y);
}
}
When two ore more points get supperposed, they are displayed with additive BlendMode.
As the number of point is quite big, it slows down the simulation a lot. I tried to adapt this code with the pixels method like bellow to speed up things :
function draw() {
loadPixels();
for(var i = 0; i < 20000; i++)
{
x = generateCoordinatesX();
y = generateCoordinateY();
var index = (x + y * width)*4;
pixels[index+0] += ???;
pixels[index+1] += ???;
pixels[index+2] += ???;
pixels[index+3] += ???;
}
updatePixels();
}
The issue is that i dont know how to generate the R,G,B,A value so that it behave with additive Blending like my first version.
Is there a way to manually calculate this ? I tried with formulas i found but after some tests, the results are different.
Thanks in advance

There's a bunch of very helpful methods that will fill in this information for you. They are respectively named red, green, blue and alpha, and you can use them to extract the relevant information from a known color.
Here's a code snippet to demonstrate:
function setup() {
createCanvas(400, 400);
background(color(0, 255, 0));
}
function draw() {
}
function mouseClicked() {
let blendColor = color(128,0,0);
let density = pixelDensity();
loadPixels();
for (var i=0; i<width*height*density*4; i+=4)
{
pixels[i] += red(blendColor);
pixels[i+1] += green(blendColor);
pixels[i+2] += blue(blendColor);
pixels[i+3] += alpha(blendColor);
}
updatePixels();
}
Once you click on the canvas, it'll update the background color by adding the blendColor to the pixel array. I deliberately choose a color that you can blend more than once to demonstrate that there won't be any stack overflow if you try to blend two colors which should go over 8-bit (255) in intensity, so you can click more than once without fearing a crash.
Hope it helps. Have fun!

Related

Using Processing for image visualization: pixel color thresholds

Image to be manipulated, hoping to identify each white dot on each picture with a counter
PImage blk;
void setup() {
size(640, 480);
blk=loadImage("img.png");
}
void draw () {
loadPixels();
blk.loadPixels();
int i = 0;
for (int x = 0; x < width; x++) {
for (int y = 0; y < height; y++) {
int loc = x+y*width;
pixels [loc] = blk.pixels[loc];
if (blk.pixels[loc] == 0) {
if (blk.pixels [loc]+1 != 0) {
i++;
}
}
float r = red(blk.pixels[loc]);
float g = green(blk.pixels[loc]);
float b = blue(blk.pixels[loc]);
pixels [loc] = color(r, g, b);
}
}
System.out.println (i);
updatePixels();
}
The main problem is within my if statement, not sure to approach it logically.
I'm unsure where this is exactly going, but I can help you find the white pixels. Here, I just counted 7457 "white" pixels (then I turned them red so you can see where they are and adjust the threshold if you want to get more or less of them):
Of course, this is just a proof of concept which you should be able to adapt to your needs.
PImage blk;
void setup() {
size(640, 480);
blk=loadImage("img.png");
blk.loadPixels();
int whitePixelsCount = 0;
// I'm doing this in the 'setup()' method because I don't need to do it 60 times per second
// Once it's done once I can just use the image as modified unless you want several
// different versions (which you can calculate once anyway then store in different PImages)
for (int i = 0; i < blk.width * blk.height; i++) {
float r = red(blk.pixels[i]);
float g = green(blk.pixels[i]);
float b = blue(blk.pixels[i]);
// In RGB, the brightness of each color is represented by it's intensity
// So here I'm checking the "average intensity" of the color to see how bright it is
// And I compare it to 100 since 255 is the max and I wanted this simple, but you can
// play with this threshold as much as you like
if ((r+g+b)/3 > 100) {
whitePixelsCount++;
// Here I'm making those pixels red so you can see where they are.
// It's easier to adjust the threshold if you can see what you're doing
blk.pixels[i] = color(255, 0, 0);
}
}
println(whitePixelsCount);
updatePixels();
}
void draw () {
image(blk, 0, 0);
}
In short (you'll read this in the comments too), we count the pixels according to a threshold we can adjust. To make things more obvious for you, I colored the "white" pixels red. You can lower or raise the threshold according to what you see this way, and once you know what you want you can get rid of the color.
There is a difficulty here, which is that the image isn't "black and white", but more greyscale - which is totally normal, but makes things harder for what you seem to be trying to do. You'll probably have to tinker a lot to get to the exact ratio which interests you. It could help a lot if you edited the original image in GiMP or another image software which lets you adjust contrast and brightness. It's kinda cheating, but it it doesn't work right off the bat this strategy could save you some work.
Have fun!

Wrong color when using loadPixels to copy image

I am trying to duplicate the ellipse such that its center point of the copy is the inverse of the original position in both axes using the pixels[] array to copy pixel data to new locations. However, the copy does not have the same color (rather than black, it is pink) as the original; the colors are not copying as I intended. Why is this?
This is the code:
function setup() {
createCanvas(600, 600);
smooth(8);
rectMode(CENTER);
fill(250);
stroke(32);
strokeWeight(10);
}
function draw() {
draw_();
}
function draw_() {
background(250);
ellipse(100, 100, 100, 100)
loadPixels();
for(var i=0; i<pixels.length/2; i++)
pixels[pixels.length-i-1] = pixels[i];
updatePixels();
}
This is the result:
According to the reference, the pixels array contains information in red, green, blue, and alpha components depending on index, like this:
pixels[0] = red(myColor);
pixels[1] = green(myColor);
pixels[2] = blue(myColor);
pixels[3] = alpha(myColor);
pixels[4] = red(myColor);
pixels[5] = green(myColor);
pixels[6] = blue(myColor);
pixels[7] = alpha(myColor);
//...
So if you simply reverse the array, you aren't going to just reverse the image. You're also going to mess up the colors: you'll swap red and alpha, and green and blue.
You might be better off using the get() function, which takes an x and y argument and returns a full color. Again, more info can be found in the reference.
Thanks a lot for Kevin's help, it's worked.
for(var i=0; i<pixels.length/2; i+=4) {
pixels[pixels.length-i-4] = pixels[i];
pixels[pixels.length-i-3] = pixels[i+1];
pixels[pixels.length-i-2] = pixels[i+2];
pixels[pixels.length-i-1] = pixels[i+3];
}

Sierpinski carpet in processing

So I made the Sierpinski carpet fractal in processing using a Square data type which draw a square and has a function generate() that generates 9 equal squares out of itself and returns an ArrayList of (9-1)=8 squares removing the middle one (it is not added to the returned ArrayList) in order to generate the Sierpinski carpet.
Here is the class Square -
class Square {
PVector pos;
float r;
Square(float x, float y, float r) {
pos = new PVector(x, y);
this.r = r;
}
void display() {
noStroke();
fill(120,80,220);
rect(pos.x, pos.y, r, r);
}
ArrayList<Square> generate() {
ArrayList<Square> rects = new ArrayList<Square>();
float newR = r/3;
for (int i=0; i<3; i++) {
for (int j=0; j<3; j++) {
if (!(i==1 && j==1)) {
Square sq = new Square(pos.x+i*newR, pos.y+j*newR, newR);
rects.add(sq);
}
}
}
return rects;
}
}
This is the main sketch which moves forward the generation on mouse click -
ArrayList<Square> current;
void setup() {
size(600, 600);
current = new ArrayList<Square>();
current.add(new Square(0, 0, width));
}
void draw() {
background(255);
for (Square sq : current) {
sq.display();
}
}
void mousePressed() {
ArrayList<Square> next = new ArrayList<Square>();
for(Square sq: current) {
ArrayList<Square> rects = sq.generate();
next.addAll(rects);
}
current = next;
}
The problem :
The output that I am getting has very thin white lines which are not supposed to be there :
First generation -
Second generation -
Third generation -
My guess is that these lines are just the white background that shows up due to the calculations in generate() being off by a pixel or two. However I am not sure about how to get rid of these. Any help would be appreciated!
Here's a smaller example that demonstrates your problem:
size(1000, 100);
noStroke();
background(0);
float squareWidth = 9.9;
for(float squareX = 0; squareX < width; squareX += squareWidth){
rect(squareX, 0, squareWidth, height);
}
Notice that the black background is showing through the squares. Please try to post this kind of minimal example instead of your whole sketch in the future.
Anyway, there are three ways to fix this:
Option 1: Call the noSmooth() function.
By default, Processing uses anti-aliasing to make your drawings look smoother. Usually this is a good thing, but it can also add some fuzziness to the edges of shapes. If you disable anti-aliasing, your shapes will be more clear and you won't see the artifacts.
Option 2: Use a stroke with the same color as the fill.
As you've already discovered, this draws an outline around the shape.
Option 3: Use int values instead of float values.
You're storing your coordinates and sizes in float values, which can contain decimal places. The problem is, the screen (the actual pixels on your monitor) don't have decimal places (there is no such thing as half a pixel), so they're represented by int values. So when you convert a float value to an int, the decimal part is dropped, which can cause small gaps in your shapes.
If you just switch to using int values, the problem goes away:
size(1000, 100);
noStroke();
background(0);
int squareWidth = 10;
for(int squareX = 0; squareX < width; squareX += squareWidth){
rect(squareX, 0, squareWidth, height);
}

How do I get a random number to stay fixed with slick?

I'm a beginner in Java as well as with the slick tools. I want to make a game that has different coloured cubes randomly placed within a certain area of the window.
I use two for-loops and call for a random number in render. I get the cubes placed exactly as I want, but the problems is that they flicker in all colours. I guess it has to do with how I call for a random number and that it gets updated with FPS?!
Please help me!!
public void render(GameContainer gc, StateBasedGame sdg, Graphics g) throws SlickException {
//set background
Image background = (new Image("res/background.png")).getScaledCopy(800, 500);
g.drawImage(background, 0, 0);
//set gamescape
blue = (new Image("res/blue.png")).getScaledCopy(20,20);
green = (new Image("res/green.png")).getScaledCopy(20,20);
red = (new Image("res/red.png")).getScaledCopy(20,20);
int xvalue = 300;
int yvalue = 400;
for (int a = 1; a < 20; a++) {
for (int i = 1; i < 10; i++) {
r = rand.nextInt(3);
if(r==0){g.drawImage(blue,xvalue,yvalue);}
else if(r==1){g.drawImage(red, xvalue, yvalue);}
else{g.drawImage(green, xvalue, yvalue);}
xvalue = xvalue+20;
}
yvalue = yvalue - 20;
xvalue = xvalue -180;
}
}
Your problem is that you generate a new random number each time you redraw the scene.
To resolve this, you may have to create an array in which you store the generated color of each cube. And each time you redraw your images, you just read each color value in the array.

Pixel reordering is wrong when trying to process and display image copy with lower res

I'm currently making an application using processing intended to take an image and apply 8bit style processing to it: that is to make it look pixelated. To do this it has a method that take a style and window size as parameters (style is the shape in which the window is to be displayed - rect, ellipse, cross etc, and window size is a number between 1-10 squared) - to produce results similar to the iphone app pxl ( http://itunes.apple.com/us/app/pxl./id499620829?mt=8 ). This method then counts through the image's pixels, window by window averages the colour of the window and displays a rect(or which every shape/style chosen) at the equivalent space on the other side of the sketch window (the sketch when run is supposed to display the original image on the left mirror it with the processed version on the right).
The problem Im having is when drawing the averaged colour rects, the order in which they display becomes skewed..
Although the results are rather amusing, they are not what I want. Here the code:
//=========================================================
// GLOBAL VARIABLES
//=========================================================
PImage img;
public int avR, avG, avB;
private final int BLOCKS = 0, DOTS = 1, VERTICAL_CROSSES = 2, HORIZONTAL_CROSSES = 3;
public sRGB styleColour;
//=========================================================
// METHODS FOR AVERAGING WINDOW COLOURS, CREATING AN
// 8 BIT REPRESENTATION OF THE IMAGE AND LOADING AN
// IMAGE
//=========================================================
public sRGB averageWindowColour(color [] c){
// RGB Variables
float r = 0;
float g = 0;
float b = 0;
// Iterator
int i = 0;
int sizeOfWindow = c.length;
// Count through the window's pixels, store the
// red, green and blue values in the RGB variables
// and sum them into the average variables
for(i = 0; i < c.length; i++){
r = red (c[i]);
g = green(c[i]);
b = blue (c[i]);
avR += r;
avG += g;
avB += b;
}
// Divide the sum of the red, green and blue
// values by the number of pixels in the window
// to obtain the average
avR = avR / sizeOfWindow;
avG = avG / sizeOfWindow;
avB = avB / sizeOfWindow;
// Return the colour
return new sRGB(avR,avG,avB);
}
public void eightBitIT(int style, int windowSize){
img.loadPixels();
for(int wx = 0; wx < img.width; wx += (sqrt(windowSize))){
for(int wy = 0; wy < img.height; wy += (sqrt(windowSize))){
color [] tempCols = new color[windowSize];
int i = 0;
for(int x = 0; x < (sqrt(windowSize)); x ++){
for(int y = 0; y < (sqrt(windowSize)); y ++){
int loc = (wx+x) + (y+wy)*(img.width-windowSize);
tempCols[i] = img.pixels[loc];
// println("Window loc X: "+(wx+(img.width+5))+" Window loc Y: "+(wy+5)+" Window pix X: "+x+" Window Pix Y: "+y);
i++;
}
}
//this is ment to be in a switch test (0 = rect, 1 ellipse etc)
styleColour = new sRGB(averageWindowColour(tempCols));
//println("R: "+ red(styleColour.returnColourScaled())+" G: "+green(styleColour.returnColourScaled())+" B: "+blue(styleColour.returnColourScaled()));
rectMode(CORNER);
noStroke();
fill(styleColour.returnColourScaled());
//println("Rect Loc X: "+(wx+(img.width+5))+" Y: "+(wy+5));
ellipse(wx+(img.width+5),wy+5,sqrt(windowSize),sqrt(windowSize));
}
}
}
public PImage load(String s){
PImage temp = loadImage(s);
temp.resize(600,470);
return temp;
}
void setup(){
background(0);
// Load the image and set size of screen to its size*2 + the borders
// and display the image.
img = loadImage("oscilloscope.jpg");
size(img.width*2+15,(img.height+10));
frameRate(25);
image(img,5,5);
// Draw the borders
strokeWeight(5);
stroke(255);
rectMode(CORNERS);
noFill();
rect(2.5,2.5,img.width+3,height-3);
rect(img.width+2.5,2.5,width-3,height-3);
stroke(255,0,0);
strokeWeight(1);
rect(5,5,9,9); //window example
// process the image
eightBitIT(BLOCKS, 16);
}
void draw(){
//eightBitIT(BLOCKS, 4);
//println("X: "+mouseX+" Y: "+mouseY);
}
This has been bugging me for a while now as I can't see where in my code im offsetting the coordinates so they display like this. I know its probably something very trivial but I can seem to work it out. If anyone can spot why this skewed reordering is happening i would be much obliged as i have quite a lot of other ideas i want to implement and this is holding me back...
Thanks,

Resources