Arvind Ganesh Arvind Ganesh - 1 year ago 131
Python Question

Writing a mandelbrot set to an image in python

I am trying to write a mandelbrot set to an image in python, and am having a problem with one of my functions.

The issue is: While I expect something like this. I am getting a plain white image. Here is my code:

Quick Summary of code:
Check if value is in set, if it is, mark it as true in an array of booleans. Then, draw the image based on the array of booleans, coloring the true, and leaving the false ones.

import math
import numpy as np
import scipy.misc as smp
from PIL import Image

def iterate(x, y, iterationNum):
z = 0
coord = complex(x, y)
for a in xrange(iterationNum):
#Don't use fabs. It can be negative.
z = z * z + coord
#This is a comparison between complex and int. It probably won't work.
#You want |Z| which is: z.real ** 2 + z.imag ** 2 > 4
if math.fabs(z) > 2:
return False
return True

def pixel(image,x,y,r,g,b):
"""Place pixel at pos=(x,y) on image, with color=(r,g,b)"""
image.put("#%02x%02x%02x" % (r,g,b), (y, x))

#here's some example coloring code that may help:
def draw(grid):
#Create a white image with the size of the grid as the number of pixels
img ='RGB', (len(grid), len(grid)), "white")
pixels = img.load()
for row in xrange(len(grid)):
for col in xrange(len(grid[row])):
if grid[row][col] == True:
#If that point is True (it's in the set), color it blue
pixels[row, col] = (0, 0, 255)
return img

def mandelbrot():
#you should probably use a square, it's easier to deal with
#The mandelbrot set fits completely within (-2, 2) and (2, -2)
#(-200, 200), (200, -200) is way too big!
TopLeftX = -2; BottomRightX = 2
TopLeftY = 2; BottomRightY = -2
#increment should be calculated based on the size of the bounds and the number of pixels
#For example, if you're between -2 and 2 on the X-Plane, and your image is 400 pixels wide
#Then your increment = (2 - (-2)) / 400 = 4 / 400 = .01 so that each pixel is 1/400th of the
#Total width of the bounding area
increment = 0.01
maxIt = 100
w = BottomRightX - TopLeftX
h = TopLeftY - BottomRightY
#This should be based on the size of the image, one spot in the area for one pixel
npArr = np.zeros((w / increment, h / increment), dtype=bool)
#Use the increment variable from above. It won't work with xrange because that doesn't
#Support decimals. You probably want to use a while loop or something
x = -2
y = 2
while TopLeftX <= x <= BottomRightX:
while TopLeftY <= y <= BottomRightY:
#I recommend using True or False in here (in the set or not)
#And then do your color calculations as I explained above
#Saves a lot of memory
if iterate(x, y, maxIt):
npArr[x, y] = True
y += increment
#once you've calculated the Trues and Falses, you'd call the draw() function
#using the npArr as the parameter. I haven't tested the code, so there may
#be a few bugs, but it should be helpful!
x += increment
return npArr

img = draw(mandelbrot())"mandelbrot.png")

I suspect the problem is with the "iterate" function in my code, because none of the values i put in iterate are returning true.

I have another issue as well, The second for loop I have here isnt even running.

Answer Source

Your handling of the y coordinate is faulty. You begin the outer loop with

y = 2

and have the loop condition as

while TopLeftY <= y <= BottomRightY:

After substituting their values, this is

while 2 <= y <= -2:

which is a nonsense. This is followed by

y += increment    

but y is already at the top end of the range. Moreover, you fail to reset y for each inner loop.

To summarise, the loop should be

x = TopLeftX                                # use the value you already defined!
while TopLeftX <= x <= BottomRightX:
    y = TopLeftY                            # moved to inside x loop
    while TopLeftY >= y >= BottomRightY:    # change the loop condition
        # ... the Mandelbrot iteration
        y -= increment                      # reverse direction
    x += increment

I am no Python expert, so there may be other problems too.

Recommended from our users: Dynamic Network Monitoring from WhatsUp Gold from IPSwitch. Free Download