Color Values, Conversion Of

Programming Questions

zebpeders..

in Programming Questions • 2 years ago

I've got a short sketch which I'm using to load an image in and then output the RGB values of each pixel. (I'm going to use MaxLink to connect this to Max/MSP ultimately, but for now I'm just working in Processing and using [println] to test the output).

Code:

PImage inputImage;

void setup ()

{

  inputImage = loadImage ("input.jpg");

  size (200, 200);

}

void draw ()

{

  image (inputImage, 0, 0);

  loadPixels();

  for (int x = 0; x < inputImage.width; x++)

  {

   for (int y = 0; y < inputImage.height; y++)

   {

   color outputColour = inputImage.get (x, y);

   println (outputColour);



   fill (outputColour);

   rect (50, 50, 50, 50);

   }

  }

}

Right now, as the loop iterates through the pixels at x, y it updates the colour 'outputColour', which for testing purposes I'm printing to the console.

The output to the console from [println] is like this:

-9473166

-9734542

-9734285

-9602956

-9537163

-9340297

-9077895

etc...

What type of data is this? Basically what I want to get is the RGB values of each pixel. Is there a mode switch I can use? These negative ints are messing with my vibe, man.

Thanks

Zeb

Replies(5)

amnon.owed

Re: Color Values, Conversion Of

2 years ago

You're kinda mixing two methods. Pixel array and get. As the reference states, getting the color of a single pixel with get(x, y) is easy, but not as fast as grabbing the data directly from pixels[]. Take a look here. It shows the basic of use of the pixel array, mainly:

int loc = x + y*img.width;
float r = red (img.pixels[loc]);
float g = green (img.pixels[loc]);
float b = blue (img.pixels[loc]);

PhiLho

Re: Color Values, Conversion Of

2 years ago

FYI, color is really just a number, combining the RGB triplet in the same 16 bits. You see negative numbers because of the alpha information on the high bits, 0xFF for opaque colors. You can also get the RGB information with bit shifts and masks.

zebpeders..

Re: Color Values, Conversion Of

2 years ago

Bit shifts and masks?

I've cleaned up the code a bit so it's just dealing with the pixel array - but I'm still getting those negative numbers. I'm comfortable iterating through and getting the data out, but it's format of the data that I'm interested in changing (from -66550651 to '120, 64, 75' or whatever the RGB may be...1206475?).

Cheers

Zeb

PhiLho

Re: Color Values, Conversion Of

2 years ago

Yes, bit shifts and masks. Examples abound in the old forum, also look at the Bitwise Operators section of the reference.

amnon.owed

Re: Color Values, Conversion Of

2 years ago

Zeb, to go from something like -66550651 to something like 120, 64, 75 you can use the functions red(), green() and blue(). As shown in the reference, the link I provided earlier and my post above. If you read the latter page, you will see that it says...

The functions red(), green(), and blue() pull out the 3 color components from a pixel.

If you want to get the rgb values from every pixel in an image you could do it like so...

PImage inputImage;
void setup () {
size (200, 200);
inputImage = loadImage ("input.jpg");
inputImage.loadPixels();
for (int y=0; y<inputImage.height; y++) {
for (int x=0; x<inputImage.width; x++) {
int loc = x + y*inputImage.width;
float r = red (inputImage.pixels[loc]);
float g = green (inputImage.pixels[loc]);
float b = blue (inputImage.pixels[loc]);
println("Pixel " + loc + " (r,g,b): " + r + ", " + g + ", " + b);
}
}
println("\nThere were " + inputImage.pixels.length + " pixels in this image.");
}

As phi.lho pointed out, an alternative would be using bitwise operators. Those are a little less intuitive, but they are faster. So once you have things running and you want to optimize, switching to those is a good option.

Top Reply