PythonPerlin

Perlin noise in python -- procedural generative art tool to seamlessly tile texture patterns in any dimensions

In 1997 professor Ken Perlin won an Oscar for Technical Achievement from the Academy of Motion Picture Arts and Sciences for his invention of Perlin noise for the Tron 1982 movie.

Installation

pip install pythonperlin

More examples and animations can be found at:

https://github.com/timpyrkov/procedural-art/

https://www.instagram.com/timpyrkov/

Generate Perlin noise

import pylab as plt
from pythonperlin import perlin

# Set grid shape for randomly seeded gradients
shape = (4,4)

# Set density - output shape will be dens * shape = (128,128)
dens = 32

# Generate noise
x = perlin(shape, dens=dens, seed=0)

# Test that noise tiles seamlessly
x = np.concatenate([x] * 2, axis=1)

plt.figure(figsize=(12,6))
plt.imshow(x, cmap=plt.get_cmap('Accent_r'))
plt.axis('off')
plt.show()

Generate domain warping

Add noise to grid coordinates and generate noise again

dens = 32
shape = (4,4)
x = perlin(shape, dens=dens, seed=0, warp=2)

plt.figure(figsize=(6,6))
plt.imshow(x, cmap=plt.get_cmap('Accent_r'))
plt.axis('off')
plt.show()

Generate octaves

import pylab as plt
from pythonperlin import perlin

# Set grid shape for randomly seeded gradients
shape = (8,8)

# Set density - output shape will be shape * dens = (256,256)
dens = 32

# Generate noise
x = perlin(shape, dens=dens, seed=0)

plt.figure(figsize=(6,6))
plt.imshow(x, cmap=plt.get_cmap('Accent_r'))
plt.axis('off')
plt.show()

# Generate noise array with 4 additional octaves
x = perlin(shape, dens=dens, seed=0, octaves=4)

plt.figure(figsize=(6,6))
plt.imshow(x, cmap=plt.get_cmap('Accent_r'))
plt.axis('off')
plt.show()

Generate water caustics

Take absolute value of Perlin noise and apply log-scaled color gradient

import numpy as np
from matplotlib.colors import LinearSegmentedColormap

dens = 32
shape = (8,8)
x = perlin(shape, dens=dens)

# Take absolute values of Perlin noise
x = np.abs(x)

# Log-scale colormap
logscale = np.logspace(0,-3,5)
colors = plt.cm.get_cmap('GnBu_r')(logscale)
cmap = LinearSegmentedColormap.from_list('caustics', colors)

plt.figure(figsize=(6,6))
plt.imshow(x, cmap=cmap)
plt.axis('off')
plt.show()

Generate flower petals

Take 1D Perlin noise as the varying radius along a circle

dens = 32
shape = (8,8)
x = perlin(shape, dens=dens)

n = 8
delta = dens
color = plt.get_cmap('tab20').colors[::-1]
plt.figure(figsize=(6,6))
for i in range(n):
    r = x[delta * i] + 1
    r = np.concatenate([r, (r[0],)])
    phi = 2 * np.pi * np.linspace(0, 1, len(r))
    scale = 1 - i / (n + 2)
    z = scale * r * np.exp(1j * phi)
    ax = plt.gca()
    zorder = max([ch.zorder for ch in ax.get_children()])
    plt.fill(z.real, z.imag, c=color[2*i], zorder=zorder+1)
    plt.plot(z.real, z.imag, c=color[2*i+1], lw=2, zorder=zorder+2)
plt.axis('off')
plt.show()

Generate vector field

Take Perlin noise as the vector angle at each point of a grid

dens = 6
shape = (3,3)
x = perlin(shape, dens=dens)
z = np.exp(2j * np.pi * x)

shape = z.shape
colors = plt.get_cmap('Accent').colors
plt.figure(figsize=(6,6))
for i in range(shape[0]):
    for j in range(shape[1]):
        di = 0.5 * z[i,j].real
        dj = 0.5 * z[i,j].imag
        color = colors[(di > 0) + 2 * (dj > 0)]
        plt.arrow(i, j, di, dj, color=color, width=0.1)
plt.axis('off')
plt.show()

Sound of Perlin noise

Perlin noise sounds nice and less buzzing than white noise

import sounddevice as sd

dens = 32
shape = (1024,)
x = perlin(shape, dens=dens)

sd.play(x, 22050)

Alternatively, save and play perlin.wav audio

import IPython
import soundfile as sf

sf.write('perlin.wav', x, 22050)
IPython.display.Audio('perlin.wav')

Documentation

https://pythonperlin.readthedocs.io