help with image

dear /prog/

an image is like a matrix of pixels

i want to blur a portion of the image, so i receive x (int) and y (int) with is the starting point of the image, then I receive width (int) and length (int) to mark the portion to be blurred.

but then the program receives also a variable called 'size' (int), which divides the portion in squares with 'size' of side. in order to blur the image, i must make the average of all the pixels in that square, and make the square with that value of color.

i already made the functions that detect the red, green and blue of a pixel [red(x,y)blue(x,y)green(x,y)] and the functions that write a color to a pixel color(r,g,b,t,x,y).

but i'm having troubles to handle the squares and the matrix of pixels, looks messy. there's like, a fuckload of fors and whiles, no?

can you help me? thanks, adojan

Get Photoshop

>>1
Have you considered KPL ?

http://www.kidsprogramminglanguage.com/kids.php

>>3

java

>>4

no, Logo is clearly the most appropriate language in this case

but i'm having troubles to handle the squares and the matrix of pixels, looks messy. there's like, a fuckload of fors and whiles, no?
Fuckload? You'd need at most 2 for loops to loop through your area. Also, putpixel-functions are pretty enterprise. Since you have an array of pixels do it inline.

function censor(image, x, y, width, height, size):
  for j in (y .. y + height):
    for i in (x .. x + width):
      bx = int(i / float(size)) * size
      by = int(j / float(size)) * size
      image.putpixel(i, j, image.getpixel(bx, by))

it looks like you're applying a pixelate effect rather than blur.
there are many kinds of blur, but think of what you're trying to do mathematically speaking, then program.
I don't know how to blur but I'm guessing you have to take a pixel, then apply the pixel color to the pixels around it in a circle fashion, with more transparency as you go away from the pixel. This is applied to all pixels.

>>6
two loops to go through the area
and two more to calculate the average of the neighborhood pixels

ONE QUESTION, WHY DO YOU NEED A PROGRAM TO DO THIS?

>>10
maybe it's cheaper and/or more flexible to use a computar instead of some optical blurring apparatus

>>1
Why do you want to censor, motherfucker? GTFO my world4ch.

http://www.imagemagick.org/script/convert.php

import Image, ImageDraw
import operator as op

def average_pixels(pix, x1, y1, x2, y2):
    pix = im.load()
    sum = (0, 0, 0)
    for y in range(y1, y2):
        for x in range(x1, x2):
            sum = map(op.add, sum, pix[x, y])
    return tuple(c / ((x2 - x1) * (y2 - y1)) for c in sum)

def pixelize(im, x, y, width, height, size):
    draw = ImageDraw.Draw(im)
    for y1 in range(y, y + width, size):
        for x1 in range(x, x + height, size):
            c = average_pixels(im, x1, y1, x1 + size, y1 + size)
            draw.rectangle((x1, y1, x1 + size, y1 + size), fill=c)

>>1 is a faggot who wants to censor vaginas

>>15

not a faggot. i'm testing the program with the head of the developers-developers person.

i just finished the image class http://files-upload.com/257911/Image.htm.html, where i have functions to aid the child class, which will have a lot of normal functions to handle pictures

>>8
yes, that is it - pixelate. basicly a section of the image is divided in squares and i do the color average with the red,green,blue function and i place it in the square

the teacher helped us with the image class, but now we are doing a child class that will use these functions, and is a bit dificult without help

adojan

>>16
Testing with head = faggot.

i'm a bit stuck still

but thanks for the help so far /prog/

adojan

Gaussian blur:
http://www.codeproject.com/cs/media/csharpfilters.asp

Gaussian blur in VB:
http://www.catenary.com/howto/gaussian.html

Gaussian blur in Photoshop:
Filter -> Gaussian Blur -> OK

function y = gaussian_blur(x)
h = fspecial('gaussian', 5);
y = filter2(h, x);

blur it along the x axis and then blur it along the y axis. result is the same, no matrices involved. PROBLEM SOLVED.

import gd and use one of the blur functions you fucking moron

GD sucks.
import Image

>>25
PIL sucks. I can't believe how popular that almost undocumented, buggy pile of crap actually is.

Script-morons don't know how to write low level libraries or even code that uses low level libraries.

>>26
Because it's fast and GD isn't?

WHBTE

troubles to handle the squares and the   total number of   iterates 624 in   the case of   MT allows one   to predict all   future iterates Combining.